Origins of the English language, a social and linguistic history

Origins of the English Language A Social and Linguistic History

Joseph M. Williams

THE FREE PRESS •^Dl

A Division of Mtictttillan Publishing Co., fnc. NEW VDRK:

^oilier MacmiHan Publishers HJNDON

Copyright €> 1975 by The Free Press A Division of Macmillan Public ing Co., Inc, All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the Publisher. The Free Press A Division of Macmillan Publishing Co., Гпс. 866 Third Avenue. New York, N.Y. 10022 Collier M
10

h

L.Lhrars of CoagreiE CHULlogisg in f uijll'.fHli)H D«t» WL11 bin г., Joseph N

Of Пи? £njjl 1ьк 1вве«вЕ*.

M S , subject and пале H i t q l , and UDrd index

[.

] .

ISBH

EniSliSn l4nJ!u4J!H --HkJioi-i'.

Title.

D-02-

I.

Sociolinj!ijiacl<:s.

Contents

Preface to the Teacher

Part I

History and Language

Chapter 1

History and Language HISTORY, HISTORIES, AND HISTORIANS. TORY AND THE STUDY OF LANGUAGE. THE AIMS OF LANGUAGE HISTORY.

Chapter 2

The Nature of Language THE BEGINNINGS. Man and Ape: The Biological Foundations, Design Features of Human Languages. Evolution toward Language. DESCRIBING LANGUAGE. W ritten Langu age. Spoken Language. GRAM MAR ANDGRAMMARS. Tradi t ional G raminais. Descriptive Grammars, Generative Grammars. GRAMMATICALLY. Usage vs. Structure. Ambiguity and Related ness. LANGUAGE AND MEANTNG. SOML CAUSES AND EFFECTS OF LANGUAGE CHANGE.

Part II

Words and Meanings Words: Native and Borrowed OLD ENGLISH AND BFFORF, The Origin, of Words. The Oldest Words in English. THE MOST

VII

11

41

Hi

le

CONTENTS

DISTANT ORIGINS: INDO-KUROPEAN. IndoEuropean > Western European. Indo-European > Western European > Germanic > West Germanic. Pre-Anglo-Sason Britain. ANGLO-SAXON BRITAIN: THE BEGINNINGS. THE RISE 01 ANGLOSAXON ENGLAND, The Christian Conversion and a National Character. The Danish Invasions. The Rise of London. CJiapter 4

From Middle English to Modern English THE NORMAN CONQUEST, THE LINGUISTIC CONSEQUENCES. Economic-Demographic Estimate!;. Political Evenln. Written English, French, and Latin in Official Texts and Documents. Written English, French, and Latin in Social Communication and Literature. Spoken English, French, and Latin on Official Occasions. Spoken English, French, and Latin in Social, Educational, and Religious Contexts. Some Additional Data Relevant to the Use of English, NORMAN FRENCH VS. ENGLISH (1066-1450). THL REBIRTH OK CLASSICAL LEARNING. EXCESS AND REACTION.

65

Chapters

The Establishment of Modern English CHANGING ATTITUDES TOWARD ENGLISH. Two Authorities. BORROWINGS FROM OTHER LANGUAGES, AMERICAN ENGLISH. Black English. American English: A Summary. SOME CONSEQUENCES OF A COSMOPOLITAN VOCABULARY. Word Families, Semantic Families.

91

Chapter 6

Creating New Words TWO GRAMMATICAL. PROCESSES. Compounding. Derivation. Grammatical Word Formation: A Summary. ROOT MODIFICATION. Clipping and Shortening. В fends Acronyms. ROOT CREATIONS. Direct Sound Symbolism. Synestiietic Sound Symbolism. Phonesthemes, PSYCHOLOGICAL AND SOCIAL MOTIVES IN WORD CREATION. OUR VOCABULARY AND ITS SOURCES: A SUMMARY.

119

Chapter 7

Semantic Change THEORIES OF MEANING. WORDS, MEANINGS, AND THOUGHTS. SEMANTIC STRUCTURE. THE RECOVERY OF MEANING. Dictionaries. Reconstruction, Citations TYPES OF SEMANTIC

153

CONTENTS

О

CHANGE. Narrowing. Some Causes of Narrowing. Widening. Transfer. Shift. Drift. Minor Types of Semantic Change, Chapter 8

Semantic Chattge in In Social Context VALUE JUDGMENTS. EUPHEMISM. Some Psycho logical Causes о f Lu phe m is m, S LAN G, A ROOT, CANT, JARGON. SEMANTIC LAWS,

Part III

Grammar and Sound

Chapter 9

Between Semantics and Grammar GRAMMATICAL STRUCTURE. IMPLICATIONS.

Chapter 10

195

215 HISTORICAL

Grammatical Change: The Noun Phrase 229 WORD OK.DLR: THE MAJOR ELEMENTS. INFLECTION: OE, ME, AND ModE, THE NOUN PH RASE: IN FLECTION S AN D STRUCTU R Б. Case, ModE and OLi. Gender, ModE and OL. Number, ModE and OE. FROM OE TO ME TO ModE. Inflections.. Relative Clauses. The Genitive, CAUSES OF GRAMMATICAL CHANGE SOME MODERN SURVIVALS OF OE INFLECTIONS,

Chapter 11

Grammatical Change: The Verb Phrase INFLECTIONS: ModE AND OE. The Past and Perfect. Person Endings. The Progressive -ing, a Problem. The Imperative and Subjunctive. THE GRAMMAR OF THE VERB PHRASE. Auxiliary Verbs. Nonstandard English. TRANSFORMING THE VERB PHRASE. Negation. Questions. Restructuring the Verb Phrase. The Impersonal. SUMMARY.

259

Chapter 12

Between Grammar and Phonology SOUNDS. STRESS.

293

Chapter 13

Phonological Change 301 TH F [ M PORTA NCE OF PRONUNCIATION. THE PHONEMIC PRINCIPLE. THE PHONOLOGY OF MODERN ENGLISH Consonants, Vowels. THE DEVELOPMENT OF WRITING. SOUND AND SPELLING. RECONSTRUCTING THE PAST, THE PHONOLOGY OF OLD ENGLISH. Vowels, Consonants.

ei

CONTENTS

Chapter 14

P!ioitola%fcifl Change; From Old English to Modern English THE NA TU RE OF SO UN D CHANGE, OLD ENGLISH TO MIDDLE ENGLISH. Vowels. Consonants. MIDDLE ENGLISH TO MODERN LNGLISH. Vowels. Consonants, SOME POSSIBLE CAUSES OF SOUND CHANGES. A NOTE ON AMERICAN PRONUNCIATION. Black English. SPELLING RLFORM.

References Index of Modern Knglish Words Subject Index

333

169 381 4П

Preface to the Teacher

Because this is primarily a pedagogical history, there are aspects of the evolution of English which I have slighted, particularly the very powerful insights into phonological change provided by generative phonology. BuL to develop the background For that would have gone far beyond what is possible or needed in a survey of this kind. I have also LtdapLed a rather conservative version of a transformational gram sn а r to deal with syntactic change. A case grammar of the son proposed hy Charles Fillraore or the model of generative semantics offered by Jami;!i McCawley or George LakolT is more interesting than the model used here. But again, the problem of striking a balance between what can be done with a class and with a seminar requires a good deal of compromise. The problems, [ would like to think, are the heart of this teAt- They can be used in several ways: 1-

23

*

A single student can teach himself the history of English if he merely reads the text and turns in the problems to be ehecked by an instructor. The text is relatively independent of a teacher. The instructor can assign some problems and entirely ignore others, The instructor can have certain problems turned in as written exercises and use the data in others for class discussions. Oft

Vtii

4, 5.

PREFACE TO THE TEACHER

He can divide the class into groups and have each group do selected problems. He can assign different problems to different students and,'or groups and have them present their solutions to the class for discussion.

The most frequent complaint will be from the student who docs not understand what "Comment" at the end of many problems requires. I am aware that students are often unable to break away from the need to believe that there arc fixed answers to fixed questions. The vagueness of "Comment" is intended precisely to force the student to pose his own questions, to overcome the intellectual rigidity that conies From high schools and junior high schools which teach that all the questions have been asked and that the student's only task is to remember the answer. He will be relieved of a good deal of his own anxiety if he is informed the first day of class that part of his task in the course is to find shape in vagueness. Incorrect approaches (or no approach) to solutions to the problems are as useful material for class discussion as interesting solutions. A discussion of why a possible solution is not a good one will get to the heart of what education is about if it centers OOl merely on *hat the beLter solution is but why it might be belter and how one might have arrived at that solution. The Czech formalist critics believe that one function of literature is to "make strange" that which we know too well. Approaching the history of language in the right way can provide the same experience. Lnglish need not be as it is now. It could have been otherwise. If some inflections had not disappeared, if more had disappeared, if the Normans hadn't successfully tricked the English at Scnlac, if we had never borrowed prefixes, if there were no such phenomenon as mL-laphor, if the Great Vowel Shift had never occurred, if any of these conditions applied to English and its history, our language would be a very different thing. And if we can bring our students to appreciate these factors, then what is like the air they breathe will become something to be aware of, to be curious about- By awakening their sensibilities, they can expand the frontiers of their imagination. Some of the more speculative questions (e.g., What would we be like if we stored memories not in the brain but in those parts of the body that had the experience?) are designed to make language itself strange. These questions, I would like to think, will permit LI moment or two of speculative play. In any event, the point of this text is to get something going in the classroom beyond the instructor telling his class what is in the text and the students dutifully writing it down in their notebooks, A criticism that will be made of this text is that it is not "tight," that there is a good deal of repetition. At least some of this redundancy is intentional. This is not л "programmed" test. But many concepts are introduced in one context with the view thai they will be useful later or because they are important enough to re-emphasize (behaviorists, read "reinforce"). I men-

PREFACE TO THE ТЕАСНЕЛ

tjo umlaut in a number of places because ii is a relatively complicated phenomenon. I repeatedly cite sociolinguistic theory and data because T want to emphasize certain concepts about language, social class, and "good grammar," Whether this repetition is finally oniy redundant dcadwood or strategic reinforcement and clarification of important concepts will be determined only at the end of the course when the student ponders how well he knows what he knows. Finally, there are a good many who have contributed to the creation of this tent. Foremost among this group arc those found in the bibliography. Without the work already done by linguists such as Otto Jespersen, Albert C. Baugh, Samuel Moore, Albert L, Marckwardt, Thomas Pyles, Fred Cassidy, Непгу'сесИ Wyld, Randolph Quirk, Fernand Mosse, and a hundred others, texts like this would never exist. More particularly, I would like to thank the students who made many valuable suggestions: Yoshi Morita, Elizabeth Temple, Mitchell Marks, U, J. Jost, and many others. I am grateful to Nancy Hall, and to Eileen Fitzgerald «f The Free Press Гог their invaluable assistance in catching many slips in the original manuscripts. Any that remain are my own responsibility. Professor William Hunter of the University of New Hampshire has my gratitude for very helpful suggestions and fur his encouragement when this text was in its formative stages. My gratitude also goes to Professor Donald Green for his help in reading the final proof. Mansct, Maine

PART I

HISTORY AND LANGUAGE

Chapter 1 HISTORY AND LANGUAGE

HISTORY, HISTORIES, AND HISTORIANS

When some people argue that histories should tell the truth, they usually mean by truth the straightforward, unadorned facts of what happened. They believe that historians who select, arrange, and shape their data to make a point about the past or who use the past to prove something about the present are not heing objective, and, hence, not telling the "truth." History, they believe, is properly the objective facts the historians can recover from the past arranged jusL as they occurred tt? relate what really happened. Such histories have never existed and never will. The individual mind of the historian, shaped by his times, influenced by his theory of history, and controlled hy his unique personal character, must always stand between the leavings of the past and the work that represents his understanding of how those leavings reveal what happened and why, Just choosing what to write about reveals the mind oi the historian evaluating and selecting from all Lhe events of the past only those most important to him. The very material he works with will become one thing for a historian who believes that history turns on the actions of great men and something else for one committed to an economic interpretation of the past- Indeed, the 3

4

HISTORY AND

modem study of history is becoming less the narratives of individual men acting in individual situations and more the analyses of broad economic and social forces that can be discovered only by highly specialised researchers, armed with sophisticated statistical and sociological tools. The intellectual focus of our age has shifted from man the individual Free agent to man's part in the overall social fabric. This shift away from the concrete individual to more abstract social forces has been encouraged by the growth of one academic specialty after another: statistics, demography, sociology, economics, psycho-history. Each has created more and more refined analytical techniques untif it is now no longer so easy as it once was for a single person to be the broadly based, general researcher who might have understood an entire age. Thus, as (he specialist has narrowed and refined his tools, so conversely have the available tools narrowed and refined the specialist. The existence of sophisticated analytical techniques pushes every investigator—historian, physicist, sociologist, literary critic to follow the methodological paths his discipline opens up to him. All researchers find not only what their age and interests urge them to find,- but what their theories and techniques predispose them to find. Not surprisingly, the reasons men write histories in the first place are as diverse as tiie theories that guide them and the methods they use. At its simplest and most elemental level, historians create a past because their community wants to know where it came from and how things got to be the way they are. The earliest myths, the stories of the most primitive peoples about their gods, are a kind tif history that explains how the uni verse and the earth came about—why the sun and the moon, why the animals, why man. The voyage of Mariner 10 beyond Jupiter and out of our solar system represents our curiosity about the origins of the universe. The Bible was once unquestioned history for most Christians. For many it still is. So once for some were the legends of the Plains Indians, and the Greek myths and Beowulf and the story of Valley Forge, Gettysburg, Lewis and Clark, and the Alamo, We no longer call all of these history because our age has different criteria for accepting some stories rather than others as satisfying ways to organize and explain our past. But the most common reason most of us would give for writing about the past is that we think we can learn from it. San tay ana's belief that those who do not remember the past are condemned to repeat it may be extreme. But it is a traditional belief. Whether or not we can in fact avoid the mistakes of history, knowing something about the past does make us understand our own condition differently than if we knew nothing about its origins. But if history is to teach us about today, the historian must arrange and shape the past to make it relevant to the present. Only then can he convince us that we should know what has happened if we are to understand what is now happening or anticipate what is about to happen.

HISTORY AND T.ANGUAOE

5

Because statements about the past so frequently touch on social values, a historian has to take a position, sometimes explicitly hut always implicitly, in regard to the values of his community, accepting and interpreting or rejecting them according to both his own personai standards and those of his time. One has only to recognize how our own history is being rewritten to include the contributions of A fro-Americans to realize how even scholarly values are changing. Only a few years ago, the intellectual-social climate did not encourage most historians to regard the Negro as a significant contributor to American history. That historians everywhere are now investigating Black history reflects a new political and social, as well as intellectual, reality. When we read any history, then, we must understand the historian's sense of what, is important about history. We must understand that the data he selects and how lie arranges them depend on his reasons for writing history in the first place, that "truth" depends finally on how his theories interpret what the interests and values of an age allow him to recogni7e as the truth. (112) British historian H. H. Carr perhaps put it most simply: . . . you cannot fully understand or appreciate the work of the historian unless you have first grasped the standpoint from which he himself approached it; second, that (he standpoint is itself rooted in a social and historical background. E. H. Carr, What Is History, Vintage Books.

HISTORY AND THE STLnV OF LANGUAGE At first glance, most of these questions may appear irrelevant, to a subject seemingly as value-free as a history of the English language A language is not a scries of spectacular events, ft was once fashionable to claim that Chaucer, Shakespeare, and Milton strongly influenced the growth and structure of English, but no longer, A language is not the product of great men acLing in dramatic situations, but rather something created and shared and recreated every day by an entire people, for the most part something shaped without any forethought or conscious planning. In fact, a history of a language is significantly different from a history of a series of clearly discrete events. The "events" in a history of the English la ng uage a re d i fficu It eve n to defi ne. They a re the seq uence of cha ngi ng abstract patterns behind the sum of countless concrete events—of hundreds of thousands of people talking roughly alike on, say, January 1, 450 A.D., in an infinitesimjjlly different way on January 2, and so on up to the present, when hundreds of millions speak billions of words every minute. A historian of English describes not how an individual speaker used language at some moment in the past, but how through time the shared abstract patterns of language have gradually changed since the fifth century A.D., when those first

HISTORY AND LAS'GUAGR

Germanic tribal from Northern Europe invaded the island of Britain and, by the mere political fact of that invasion, thereupon began speaking a languajje we no longer call West Germanic or Frisian or Jutish or whatever, but Old English (which we shall henceforth frequently abbreviate as OR). Traditionally, historians of language have studied three kinds of patterns or structures in language: The sound patterns, or its phonology; the syntactic and inflectional patterns, or its ктащгпат: and the patterns of meaning, or its semantics. We could study these aspects of Rnglish without ever referring to a single person or historical event. We would then have an internal history of the language, a history of changing structures and patterns with no reference to the extern a J history of the language. The external history of a language is partly defined by a list of events which influenced masses of speakers to change those phonological, grammatical, and semantic patterns which make up its internal history. So in that sense, the internal history of a language determines what we choose to call its external history. The major events in the external history of Rnglish are easy enough Lo list: С 450; c, 850: 1066: 1204: С 1348: 1476: 1607: с 1945:

The Northern Germanic tribes invade and occupy Britain, The Danes invade and occupy the north and east of England, The Norman French invade and conquer England. The Normans in England are forced to choose England or Normandy as their homeland, The plague wipes out one-third of the English population. Printing is introduced into England. A meriea i s со Ion i zed by Engli sh s pea kers. American political and cultural dominance makes American English a candidate for a world standard.

But, in addition to these major events, are less easily defined pressures exerted by the changing social class structure that began to replace the rigid feudal system of pre-1200 England, social pressures which ever since have shaped and reshaped our speech. Since all these events have caused our linguistic patterns to change, we would want any history of the Lnglish language: (1) to describe English structure at various critical stages of its development, and (2) to explain why one form of the language was replaced by another through the social forces set in motion by those large and small events.

THE AIMS OF LANGUAGE HISTORY All of this seems to be straightforward, value-free history. But if we conclude that writing a history of a language is divorced from the kinds of problems a

HISTORY AND LANGUAGE

7

historian faces when he write* a social or economic history of, say, Ihc postW'WI I period, we would be quite mistaken. The same kinds of problems about theories of history (and language) determining a historian's questions, about his intention determining his selection and arrangement of data, about the values and social uses of linguistic history—all should con From the historian of language no less severely than they do the historian ofpolilics and wars. For example, almost all histories of English ignore the development of forms such as these: He gone now, ain't lie?

My frien' done

He always be late.

Until recently, such sentences were thought by many grammarians {and most teachers) to represent an ignorant, illiterate, lazy, uneducated, probably unintelligent and certainly ungrammatical English. It is the English of the lower classes, stereotypical!у Southern, often Black, a dialect of English traditionally beneath intellectual contempt. But for many obvious social and intellectual reasons, linguists have recently begun to etplon: much more seriously than before the patterns of "nonstandard" English in general and the English of certain Шаек Americans in particular Because these researchers have also found new investigative tools in modern linguistic theory, they have decided that such sentences are not ungrammatical in the sense that they have no predictable regular pattern. Quite the contrary. They have a complex, orderly structure that has just never been sought, discovered, and incorporated into grammar books meant for those being taught the prestige dialect of our culture—except as examples of what to avoid. No one really knew what the structure of nonstandard Knglish was becajsc few researchers ever bothered to study it. But in the last decade, as our attitudes toward minority groups in this country have changed, as more and more " nonstandard" speakers have confronted a "standard"-sneaking society that for decades has kept them out of the colleges and the centers of social and political power- as new linguistic theories have become available, the historian of the English language has begun to look for—and find—new " f a c t s " about the history of nonstandard English, facts that exist only as a consequence of a changing intellectual climate and a new body of linguistic theory. This text will deal with these questions because its author, influenced by his time, finds them intellectually challenging and socially important. Language varies through time, largely because it varies through social space. Lower class, lower-middle class, and upper-middle class speakers interact in a variety of ways. One way that crucially affects the development of language is how the speech patterns among those groups differ and the way each group responds to the other's speech. We choose one pattern of speech rather than another not by some immutable rule of logic or history, but by noting how those who wield social, political, and economic power habitually

If

HISTORY AND LANGUAGE

speak. If we attempt to rationalize why we choose to speak in one way rather than another, we may use the history of English, logic, or dogmatic opinion to justify our usage. But how we speak—whether we say ain't Ior area'! / o r am I not—finally depends not on rational, logical reasons but on the (usually) unconscious linguistic preferences of those who command our rcspcet and who, by their prestige, silently influence the writing of textbooks and the cond uet of ed ucati on. in some small way, then, perhaps an understanding of the historieal sources of both standard and n on standard dialects and the social base* by which we decide about standards might qualify some of the historically unjustified responses many of us have toward those who use "imgrammatical, illiterate, uneducated speech." More significantly influencing the nature of this particular history, however, are its larger pedagogical goals. Most histories of English seek only to summarize the outstanding ways in which the language has changed and some of the causes for those changes, LO comment on the rise of dictionaries and changing attitudes towards standards of usage, and to suggest the range of language diversity across geographical space—all presented for the reader's information. If the student succeeds in com mitt ing the facts to memory, he has satisfied his responsibility toward the subject and the book. The intention and design of this hisLory arc different. It will try not only to transmit information about the history of Lnglish but also to suggest in some modest and very incomplete way what historical linguists do when they study and describe the history of a language. It is not a history to be read passively, it will ask the reader to engage himself with historical data presented as problems, in most eases, a discussion follows each problem, so it will be easy for the student to ignore the problems and read just the discussion. 1 can only urge the student who wants to do more than memorize facts to wrestle with the problems before he goes on to the discussions. The bibliography lists several excellent histories of English that, summarize and present many facts abouL the language in an interesting and scholarly way. For the reader interested only in facts, I recommend them. (7, 58, 95, 172, 182, 209, 244) This history will ask the reader to create much of the history of the language himself. It will force him to become as much of a historical linguist as can practically be managed in a book where all the relevant data has already been culled and the questions at least tacitly formulated. In addition to acquiring some sense of an intellectual discipline that has its roots in Plato's Cratylus and beyond, and of some of the particular fruits of that discipline, the student will, I hope, also come to understand something about language in general. I have already mentioned hnw important our speech patterns are in our social dealings with one another. We identify ourselves as members of a geographies! dialect, a soeial class, even a particular age group by our specuh. But our language is unique in the animal kingdom. We do not exaggerate

,

HISTORY AND LANGUAGE

9

when we say that nothing defines man better than the fact thaL he is only the animal that uses language creatively. He is the only anima! that can create out of a finite number of linguistic units a potentially infinite number of sentences that can correspond to the potentially infinite number of situations lie faces in the world: Lake Erie is polluted. and, even more importantly, to the equally large number of conceptions that do not yet correspond to the real world: If we continue to pollute the oceans, maybe they wJll die, and, indeed, to those that can never correspond to reality: The rectangular onloseny unfurled its insolent symmetry. Finally, some of the problem* will ask the student to exercise his linguistic imagination to create new words, new meanings, new grammatical patterns, new metaphors, to explore new semantic spice which his mind must create. One of the current commonplaces of education is that our language creates out world. It provides us with the categories of our experience, and in но doing closes off from us other ways of structuring our cogniLive world. So strongly stated, such a claim is wrong. But by flexing his verbal imagination, by creating new linguistic patterns, the student may become more sensitive to some of the creative possibilities of language. These problems have no right or wrong answers. Finally, much of the data here has been over-simplified, an unavoidable consequence of trying to present a unified view of what is in reality often diverse and contradictory data. Regional and personal variations, inconsistent historical developments, have now and then been ignored in an effort to make the his Lory of the tnglish language coherent enough for one to grasp at least the broad outlines of 1500 years of change. And yet it is probably that wrestling with inconsistency and variation that most characterizes the historical study of
Chapter 2 THE NATURE OF LANGUAGE

4 THE BEGINNINGS

Л story often cited to illustrate an idea which some have held about the origin of language is the one Herodotus relates about King Psammetichos of Egypt. To determine whether the Kgyplians or the Phrygians were the older race, he ordered iwu children to be raised entirely isolated from speech. He believed one would, quite on his own, begin speaking one of the languages as the original Ur-language of man. When one day one uttered the sound btkos, which meant bread in a dialect of Phrygian, he concluded that Phrygian was the older language. An early Scottish king who performed the same evperiment concluded that the first sounds the isolated children uttered were pure Hebrew. This idea that our language has descended from some older, probably " p u r e r " and more fundamental source is reflected not only in the Roman belief that Latin was corrupt Greek, but that the babel of languages afflicting the human race results from the confusion visited upon the one original, pure language spoken by the builders of the tower of Babel. The belief persists even today: At some earlier time, language reached a

12

HISTORY AND LANGUAGE

state of perfection; since then, it has been all decay and corruption. Some have claimed that English is particularly degenerate, for "it has no grammar" to speak of in comparison with the rich inflectional systems of older languages such as Latin and Greek. English inflects nouns for only two cases: genitive— the boy's hai, and the general case, distinguished hy the lack of any endings— the boy. Latin lias six inflectional cases: nominative, accusative, dative, genitive, ablative, and vocative- Sanskrit, the oldest fully recorded known language related to English, had eight cases; nominative, accusative, dative, genitive, ablative, vocative, locative, and instrumental. Obviously, then, English is in the final stages of a decay that began at least 3000 years ago. In a few hundred years perhaps, even the inflected genitive in boy's will disappear, along with the third person singular s, the past tense -ed, and the comparative -er and -ей endings, leaving us with the rotten hult of a once richly grammatical language. If we want an authority to decide a problem of grammar or usage or meaning, then, we should (it is claimed), loot to the past, usually to Latin, less often Greek: One does not end sentences with prepositions in English because one does not do so in La Lin. Nor does one ask Who did you see? because as the object of see, who should be whom, the accusative case. Case agreement, of course, is required in good Latin, (45, I S3) Unfortunately, the real origins of language are as completely unknown as its evolution preceding the last four or five millennia. Assuming we can reject the idea that language was a gift of the gods, we have to ask how it was possible for a hominid that presumably communicated not too much differently from our present primate relatives to evolve into a creature who can do what you and I are now doing. In the nineteenth century, speculation about the origin of language was so widespread in the scholarly journals that in 1866 the Societcde Linguistique de Paris finally had to ban the publication of any more articles on the subject. Then for about a century, the question was unfashionable. Л few eminent scholars speculated about the origin and evolution of language, but it was not a problem actively pursued. In the last few years, however, the subject has once again become a live issue, particularly among anthropologists. The problem can be broken down into at least three parts: 1. 2.

3.

What are the biological differences between the neurological and peripheral speech mechanisms of man and other primates ? What are the important differences in the "design features" of the language of шал and other creatures and how can they be related to the biological differences? How could the biological and design features have evolved from those possessed by a creature that was the ancestor of both ape and man?

THE NATURE OP LANGUAGE

13

Man and Ape: The Rit>lc»«k-ii 1 Foundations

The biological question is a very difficult one to answer because so little is known about the neurology of speech in man, much less the neurology of communication in apes, and because what is known sheds very little Sight on the question or \i of debatable reliability. There arc some obvious differences between man and ape. Man has a higher brain. But this alone is insufficient to explain the capacity for language. There are creatures known is nanoccphalic or birdheaded dwarfs, humans who arc normally proportioned but who never grow more than two-and-a-half or three feet tall. Their brains average perhaps .4 kilograms compared to an average adult's I -35 kilograms. Some femaie chimpanzee brains weigh as much as .45 kilograms, larger than the brains of many nanocephalic dwarfs. Yet while the dwarf, though often retarded, usually speaks as well as a five- or six-year-old, the chimpanzee never speaks. Nor does the ability to speak lie entirely in brain-to-body ratio. A human, adult brain-to-body ratio is about 1:50 or so. The adult chinip ratio is 1:100 or so, but the adjk rhesus ratio is i :40. There is more total surface area in the human cerebral cortex because of its extensive folding. But it seems fairly clear that brain size alone is insufficient to explain language capacity. (130) It has been claimed (and denied) that the neural anatomy of man's brain is fundamentally different from an ape's in at least three ways. First, man's brain is latcraiized: That is, certain cognitive and motor skills have localized in either the left or right hemisphere. Нот most righthanded people, the Jeft hemisphere of the brain controls manual skills and language (and vice versa for most left-handed people). Although apes often seem to prefer one hand over another, suggesting a potential for lateralization, it does not seem to be as profound a tendency as in man. The fact that manual skill (i.e., the skill associated with tool-making and tool-using) is usually localized in the same hemisphere as speech has led some anthropologists to speculate that tool-making either necessarily preceded or developed concurrently with language. Since the most recent finds in Africa suggest that the earliest tools date from perhaps two-and-a-half million years ago, it may be that man has been taiking for about that long. But even if this is the case, it is very difficult U) understand why speech should depend on la legalization. Furthermore, there is abundant evidence to show that if the left hemisphere is severely damaged in very young children, the right hemisphere wili take over the neurological speech functions. Thus it is not merely la Legalization that is responsible for speech in man. Second, it has been claimed (and denied) that there is an anta in the human brain, the angular gyrus (see Figure 2.1) that is not found in other primates. In man, no neural pathways directly connect the visual, auditory,

14

HISTORY AND LANGUAGE

Figure 2.1. The Human Cerebrum and somesthetic (tactile feedback) areas. Rather, neural pathways from each of these areas are mingled in the angular gyms, creating a neural switchboard of great complexity, one which perhaps allows us to coordinate the ability to utter (requiring fine motor control) a sound (requiring aural feedback) that becomes the name of something seen (a visual experience). Apes allegedly do not have such an area, a claim somewhat supported by the observation that they do not learn to obey spoken commands very readily, but do respond to visual cues. Third, it has been alleged (and denied) that man's linguistic system is more independent of his Liiubic system than that of the apes. (The limbic system is that inner patt of the brain and brain stem which is in evolutionary terms, much older than the cerebral cortex and which controls basic motivation, emotions, the more primitive mental functions.) What this is claimed to explain is our ability to verbalize independently of our emotional needs. Apes appear to communicate only when their emotional needs are involved: se\, social contact, fear, hunger, and so on, perhaps because their communication system is more directly linked to their limbic system. We need not be sexually aroused to make noises about it, nor need we make noises about it when we are. (123) Finally, it is clear that our peripheral speech mechanism is substantially different from the ape's. Although the same gross anatomical features can be found in both man and ape: lungs, diaphragm, larynx, glottis, tongue, teeth,

THE NATURE OF LANGUAGE

15

lips, and so on, in the ape, they are not integrated into a speech system. Even in man, none of these features was originally intended specifically as part of a communications system. Each feature lias its own biological function entirely independent of talking. But in the ape, those organs still serve almost exclusively their original functions: breathing, swallowing, dosing off the lungs to prevent the escape of air, chewing, and so forth. The laryngeal structure of apes simply does not allow them to articulate human sounds. In fact, it is for this reason that a number of researchers are now attempting to teach, chimps language through some medium other than verbal: sign language, plastic chips, computer panel displays, for example, (68,168, f 86) So along with the development of brain anatomy, we have to explain how a set of independently functioning organs came to be integrated into a physiological system connected with a cognitive system. Presumably, it did not happen in one great leap across genetic space. Presumably, environmental pressures selected those creatures who were best able to use even the beginnings of.such a system toward an end that enhanced their chances for survival.

Design I'VLII иггч of Human Language's

Keeping these very tentative biological possibilities in mind, we might now ask what the crucial design features of human language are and whether any of them depend on what is biologically unique in man. One set that has been proposed is the following (87,88): 1.

2.

3.

4.

5. 6.

7.

Vocal-auditory: Although other channels may be used (sign language or braille, for instance), human languages are fundamentally spoken and heard. Broadcast transmission: The signal is ordinarily transmitted in al! directions simultaneously rather than through a narrow channel (the difference between a smoke signal and spelling letters in the palm). Rapid fading: The signal does not remain in the channel. When a word is uttered, it disappears from our perception virtually instantly, unlike the lingering chemical traces of dogs or ants. inkTchangeiiMiity: Ordinarily, all normal users of human languages can both transmit and receive, unlike many insects whose direction of communication is sex-determined. Complete feedback: Normal speakers are aware of their own message. Specialization: The energy needed to communicate is less than the results intended by [he communication. One need not cut down a tree to Communicate the idea. Scjnanticity: The signal has meaning. It regularly correlates in some sense with that which is within the perceptual or cognitive universes of the sender and the receiver.

16

8.

9.

HISTORY АГ'О LANGUAGE

Arbitrariness: The signal bears no iconic relationship to thai which is communicated. The angle of a bee dance relative ю the sun directly represents the direction of the food source from the hive. The less intense the dance, the farther the food from the hive. The sentence, The food is в mile north of here docs not represent distance or direction i conical!}1. Discreteness: The message has parts. Each word in this sentence discretely represents a bloc of information that combines into larger structure*, and those structures into larger structures (phrases and clauses). The three distinct warning cries of a vervet monkey (one for overhead danger, one for large-creature ground danger, one for snakecreature ground danger) holislically communicate each message. They have no decomposable inner structure that could be represented by discrete bits of information; overhead

"I

ground П - J l <•«•* 10. Displacement: The message can be com mu nicated at a d i sta nee from Ih at which it refers to. Humans can speak of ancient Egyptians, who are displaced in both time and space. The ability of a non-human to " h o l d " the message associated with a stimulus is distinctly limited. 11. Openness: New messages can be constructed about any aspect of experience. Bee dances concern only food and the location of new hives. They do not (cannot) "talk about" the weather, what is happening back at the hive, and so on, 12. Tradition; The particular forms of particular human languages are passed on culturally from generation to generation. Bee languages are passed on through the genetic code. 13. Learn ability: The particular forms of particutar languages can be learned. Bees from one species cannot learn to communicate with the bees of another. (Certain birds, however, do acquire the "dialects" of other binds of the same species. When neighboring flocks have slightly different territorial songs, it is possible for them to "learn " Lhc others' dialect.) 14. Prevarication; Human speakers can consciously construct messages a bout that which is counterfactual. Some primates may also have this ability, 15. ReHexivcuess: H umaiv^ can use language to talк about)anguage. 16. Duality of patterning: Human communication is materially structured on a phonetic level through the discreteness of a finite number of discernible sound classes, on a. grammatical level independently of the phonetic level through part-of-speech, subjecL-verb structures, and so on, and through an associated semantic level that is partially independent of the grammatical level.

THE NATURE OF LANOUAnfi

17

More recently, some linguists have claimed additional universal characteristics of human languages. (31) Among them arc the following: 17. Deep-surface structure: Sentences arc organized on two levels: Their surface level, Lhc observable order of parts of speech, and so on, and a deep level, the underlying grammatical relationships, (This is dealt with more fully later.) 18. Grammatical categories: ЛИ human languages have grammatical categories which can be gencraliy covered by the terms substantives (or nouns) and verbs. Substantive-verb relationships include actor-action, recipient-action, instrument-action, and a few others. When we concentrate just on the ability of speakers to use language, however, we bring out what may be the most important differences between man and his relatives, making the problem of an evolutionary accounting yet more complex. 19. Non-instrumental communication: The message may be freed from any immediate consequences. That is, non-human communications seem always intended to bring about or maintain a state of affairs involving territorial claim, danger, mating, food, or social status. But a human can utter a sentence like / wonder who's kissing her now without wanting to bring about a new state of affairs. A further consequence of this is that human messages need have no audience to validate them. Animal signals usually do. 20. Message isolation: The message may be communicated independently of the communicator. Even among the most advanced primates, message forms ordinarily consist of not only $ound, but a complex of features including facial expression, posture, eye fixation, distance from receiver, and so on. Except for agonistic cries, primates in their natural state usually require a multi-modal transmission of a message. Man can communicate by telegraph, which screens out virtually all paraiinguislic signalling, 21. Cross-modal communication; The message may be detached from a sign in one channel-type and attached to a sign in another; i.e., we can tap out morse code, read signal flags in reply, and relay the message in braille. 22. Meaning isolation: The meaning of a message may he detached from any overt articulation of Ihe message. Thai is, we can (though we need not) think in language without any obvious enervation of muscles controlling articulation. 23. Meaningful violation: Human speakers n a y consciously violate rules governing the struct иге of a message, and their audiences can impose an interpretation on them; c.c. cum m ings' anyone lived in a pretty hun?

IS

HISTORY AND LANGUAGE

(wlik up so floating many bells down) docs not elicit a blank uncompreh ending stare. But if a mutant bee began doing his own dance, the other bees would be unable to impose an appropriate meaning on it. 24. Sign change: The form of a word may be separated from its original meaning to take on a new meaning. The word grasp, for example, has transferred from the manual to the cognitive sense. Similarly, the meaning of any word may change or the word may disappear entirely: Meat once meant any food; score, an Old English word for a garment, has disappeared entirely.

Involution toward Language Now the great problem with all of this is to synthesize what we knowabout the biological foundations of language with which of these features we decide must have been among the first steps toward language. We might reasonably assume that the peripherat physical anatomy evolved as a consequence of selection for greater cognitive language ability. Ц seems difficult to understand how the diaphragm, larynx, tongue, etc, could integrate into a speech system before the neural groundwork had developed in the brain. The next question is whether the brain changed in one massive genetic mutation or whether different parts of the brain slowly integrated into a linguistic neural system in perhaps the same way the peripheral speech system must have slowly Integra Led into what it is now. The crucial innovations toward a language that we would call rational are displacement and openness. The others—duality of patterning, semaniicity, broadcast tninemission, and >o on—either necessarily follow these two or obviously precede them. As a category, displacement includes the ability to verbalize or not in the presence of a stimulus that might ordinarily elicit verbal behavior: fear, hunger, sexua! attraction. The corollary is the ability to produce the signals associated with those stimuli when they are not. present. But uttering a signal of some sort when no gratification is involved is such startlingly innovative behavior that it demands a preceding change in neural organization. If indeed we can use our verbal system independently of our limbic system (as apparently we can}, then we must postulate a mutation that has had major consequences: A food cry thai could be produced when no food was in the immediate perceptual environment could eventually become a signal cry for an intention to find food, and then the " n a m e " of food. This kind of displacement ultimately allows us to hypothesize, to lie, to wish, to predict, to contradict, to contradict a contradiction, and so on, for all of these are speech acts independent of any immediate physical referent. At the same time, however, the other characteristic, openness, the ability to combine a finite repertoire of signals into novel sequences with novel meanings, also had

THE NATURE OF LANGUAGE

19

to develop. Otherwise, we could not now wish Foi1, predict, or lie about that which never was. It is the combination of these two unique abilities: displacement, which allows us to verbalize about that which is not immediately present either in our environment or in our internal emotional state, and openness, which allows us to create completely new messages never before uttered, that is al the has is of rational language. For if either of these are lacking in a creature, we do not attribute to him thetapacity for rational conversation. Almost certainly, the hominid thai was able to use and understand the greatest number of signal cries was best able to hunt, male, and survive. At some moment, perhaps, one such creature must have uttered two signal cries in succession to create what was, in effect, a new message never before uttered. But now we face the additional problem of explaining how his audience was able to understand such a radical communicative innovation. For any advance in communication requires two concurrent abilities: The ability to transmit by means of the innovation, but also the ability to receive and understand it. Now, what is known about the ecology of our simian ancestors that might have created the selectional pressures leading to openness and displacement? Because their period of existence has repeatedly been pushed back farther and farther into the mists of pre-history, explanations based on changing environmental conditions in any given era must be suspect. For a time, it was believed that our distant ancestors descended from the trees when a climate change caused their forests to disappear, leaving them grasslands as a new habitat, a more differentiated world than their forest. Because more open spaces may have required a greater capacity to deal with new visual experiences, visual sensitivity came to be much more highly valued. Perhaps (a favorite word, along with possibly and the phrase it may have been, when anthropologists come to the crucial moment, in their argument), perhaps those hominids needed a great many more signal cries in the grasslands than they originally needed in the trees. Finally overburdened with loo many discrete cries, even while larger brains (i,e., larger cerebra! cortexes) were being selected for memory, a creature or creatures "hit" on combining cries as the answer. As the brain grew, its organization changed, allowing the cries to be used in contexts where they were not intended as immediately relevant signals. Perbapsas the cerebral cortex grew, it also assumed a larger share of control over behavior, replacing the limbic system to a greater and greater extent, allowing those creatures to utter a signal independently of limbic involvement. One possible contevt in which such signal cries might have- been used without iheir "referents" being present is play. If the young of »ur hominid ancestor played as the young of all primates play, they "acted out" a good deal of adult behavior. Part of that behavior may (Note the equivocation) have included play hunting, with appropriate signal cries. (No reader should be unaware of" the leap through evolutionary space in that sentence.) The

20

UI5TOE.Y AND LANGUAGE

brightest of the young were thus able to use language independently of its primary stimuli, perhaps those who played the longest into young adulthood. The fact that the young of humankind is so dependent for so long makes them tin usually "plastic" in their development, unusually susceptible to environmental experience as opposed to innately ingrained behavior. The whole question has such a fascination that one understands why so many scholars want to propose their pet explanations. One should not wax too metaphysical over every pulling biological phenomenon. But, along with life, sentience, and self-consciousness, human language may have to be included among those features of creation which arc built on principles of organization that transcend our ability to understand, much less explain. In any event, how man progressed from even this primitive poLemiality for syntax to the highly inflected languages of Sanskrit, Hebrew, Greek, and Latin is similarly an object of pure speculation. But whatever their origin, to get back to the original point, there is no a priori reason to argue that this evolution inevitably led to extensive inflections as the highest form of linguistic organization. It may well be that languages develop through cycles, at one time relying on inflections, then losing them as they come to rely more on word order and phrasal ways to express ideas, then developing new inflections out of words closely associated with other words, then losing inflections again. But.this theory covers such a vast period of linguistic evolution that there is littlesohd^evidence to confirm it as ^universal oT language history] Probably the main reason many do believe that the complexity of a grammar depends on its inflections is that the first grammars, the grammars of Sanskrit, Greek, and Latin, were written for highly inflected languages, languages which we now esteem because in them we find much thaL is at the foundaLion of Western and Middle liastern civilization. These languages became so intellectually prestigious that for centuries, they determined how scholars thought about Lhe grammar of al! languages. Yet, as we shall see, English, which lacks an elaborate set of inflections, still has a grammar so complex that we shall scarcely be able to give it more than a superficial examination.

DESCRIBING LANGUAGE The question of where to look for the primary data for a description of linguistic structure, for the derived data of our history, is a crucial question. We have already mentioned the popular idea that only upper-middle class educated speakers have a grammar worth studying. And because the most presLigious representation of that prestigious dialect is usually in its written literature, many believe that writing is where the real grammar of a language can be found, that speech is merely corrupted writing. This encourages still

THE NATURE OF LANGUAGE

21

more those who would look to the past for grammatical respectability, because all we have left from our distant linguistic past is writing. And ordinarily, all that is preserved are the most important and hence most self-consciously composed documents: laws, religious writings, and so on.

Written Language The question of whether to begin with a description of written Latin or Old Fngh'sh or whether to construct first a description of contemporary English against which older forms of English can be contrasted might be put as follows: Г Г we want a theory of language that forces us to ask more rather than less demanding questions, would such a theory emerge if we based it on what we know about the language of texts in a dead language or on what we know about our own, living language ? What is. left from dead languages is like a drop of water from an octan of speech. The only concrete evidence of how language was used in the past is from what those speakers happened to write down and what the viscissitudes of time have spared. But even if a theory of langjagc were based on written material from modern languages, it would still ignore some of ihe most crucial facts that uniquely characterize the way humans use language. Writing is a comparatively recent invention, only about 5000 years old. Modern man has been around for at [east 200,000 years, perhaps several times that long, and certainly talking the whole time. In fact, when we consider that a minority of just the some four to five thousand languages spoken today have ever had their own written form, we realize that writing is a relatively rare mode of communication, that long before writing existed, thousands of languages must have developed, flourished, and died unrecorded. More importantly, writings that have survived from the distant past of English (more than 1000 years ago, let us say) do not give us a clear record of the language used in ordinary, everyday communication by the great mass of people. Reading and writing were--and in many cases still are- arcane skills known only to a relative handful of educated men. Yet, while writing inevitably has greater prestige than speech, we know that writing changes because speech changes more often than speech changes in response to writing. We also know that languages change because of the ways prestigious and поп-prestrgious forms of speech influence tine another. Many features of what we would now call "grammatical" writing certainly had their source in what once would have been called "ungrammatical" speech. Thus it" we relied only on evidence from the past to construct a general theory of language, we might seriously underesiimale the diversity of dialects spoken within a single language. We might possess documents reflecting

22

HFSTORY AND LANGUAGE

different geographical dialects. But wt would not have records reflecting the speech of the ordinary speaker, or, for that matter, even the casual speech of the person who did the writing.

Spoken Language Another aspect of languge that writing preserves at best imperfectly is the way a language sounds. Much has been made of" the fact that humans vocalize their language. In fact, one common definition of language among the school of linguists called descriptivists it that it is a set of arbitrarily structured uttered sounds. Modern linguistics in the first half of this century strongly emphasized this aspect of language for two reasons. First, they inherited a rich tradition from nineteentlvcentury philologists who had achieved great success in establishing through parallel sound patterns the fact that the Germanic languages (English, German, Dutch, and the Scandinavian languages), the Romance languages (Latin, French, Spanish, Italian, Portuguese, Rumanian), Greek, the Slavic languages (Russian, Polish, Czech, and so on), Persian, the ancient language, Sanskrit, and snme modern languages spoken in Northern India were all descended from a hypothetical common ancestor, Indo-European. The detailed analysis of sound changes showed, for example, that English brother\ German bruder* Latin/rarer, Greek pkrater, and Sanskrit bhraatr were cognates, or words desccndixi from an earlier word that can only he theoretically reL-onstructed as *bhrater (the * indicates that the word has never been found in any recorded text, that it is a reconstruction), (t was one of the great intellectual achievements in the history of language SLudy. (163, 183) Second, and more importantly, in the second quarter of this century, a new breed of linguists emerged from an anthropological discipline built around the questions anthropologists faced when they first walked into a village that spoke a language completely unlike any of the Indo-European languages. Before they could begin their research, they had to decode what they heard. Because the first aspect of the language accessible to them was its sound, descriptive linguists from the twenties to the present have labored to devise ways to discover and describe sound patterns. They were also influenced by an auiLudc among other social scientists that valued most highly that data which was objectively measurable. (235) As we shall sec, because linguists have devised many ways to recover from writing the general outlines of past phonological systems, we are not completely cut oft" from the sound patterns of a dead language. But the imperfect relationship between spelling and sound in English should make it clear how difficult it is to use writing to discover pronunciation. One aspect of sound that is almost completely lost, however, is the rising and falling of our voices in questions, statements, etc, the different ways we

THF- NATURE OF LANGUAfifc

23

accent one syllable rather than another as in PERveri and perYtRT, John ALsofound some money and John also FOUND some money. Again, there are ways to approximate what these suprasegmenfals may have been [ike—particularly through the meter of poetry— but these ways are imperfect at best. And of course, it is just about impossible to determine what sorts of vocal signals speakers of an extinct language used to ex press surprise, anger, bewilderment, sadness, and so on. We cannot assume that all speakers of all languages use the same signals to indicate these emotions, for jiaralinguistic features of language vary froni culture to culture. So in addition to excluding much of the variation of language across social, stylistic, and geographical boundaries, writing also partially cuts us off from the patterns of its sounds. Relying on the past to construct a theory of language would severely limit the power of our theory. 2.1: Man could conceivably have developed a communication system based on any of his senses: sight, smell; feeling, taste, hearing. Assuming that evolution tends to work towards the most efficient biological system possible within the genetic limitations of the organism, how would you justify sound over any of the other sensory systems that might have evolved to an extremely high level of complexity as a basis for communication? That is, assuming that our sense of smell might have evolved to a very high degree, what disadvantages would it still have in comparison to hearing ? PROBLEM

PROBLSM 2.2: Review the design features and capabilities of human languages. Then outline a science fiction plot in which humans confront a society that lacks any two of the features. For example, what would happen in a society which communicated by producing balloons with sentences spelled out, as in a comic strip? And suppose those balloons did not have the characteristic of rapid fading but rather dissolved over a period of an hour or so. Where would choral societies have to practice? How would secrets be transmitted? What would Hamlet feel like? What would Beethoven's Ninth smell like? What would Homer taste like V 2.3: What features of language can writing communicate which speech cannot4.

PROBLEM

CKAMMAR AND GRAMMARS

More about a dead language can be found in its written form when we investigate its grain mar, O r the ways words are made up out of smaller pieces

24

H1STORV AMp

GE

and then put together into larger phrases and clauses, the way unhappiness h made up of un-, -happi-, and -леи, the way a sentence like Unhappsness can be an educational experience can be structured: SENTENCE SL'EJECTVNOON PHRASE

PREDICATETVKRIS PHRASE AUXILIARY

VERB

NOUN un-

-haf>pi

COMPLFCMENTVNOUN PHRASE ARTICLE

-ЛЕИ

can

be

an

ADJECTIVE

NOUN

edwea- -Поп- -al

But now a multitude of questions arise when we ask what is involved in formulating the grammar of a language we all speak, much less one we don't.

Traditional Grammars Traditionally, writing a grammar has meant first identifying parts of speech, usually by means of semantic definitions: А лоно, Wkcbny, is the name of a person, place, or thing; an adjective modifies or restricts the idea, of boy: big boys; a verb shows action: big boys play. These parts of speech are then placed into paradigms, or lists that show the inflectional endings they occur before: boy-boys-boys-boys1; wolk-waiked-walked-walk'mg, for example. Even though English has a very limited paradigmatic inflectional system, traditional grammarians have emphasized these patterns and their meanings (past, future, present, perfect, progressive, and so on, for verbs; singular, plural, possessive for nouns) because grammars of highly prestigious languages like Latin and Greek depend heavily on elaborate inflectional paradigms. Without considering whether a new conception of grammar might be appropriate for Lnglish, the least, enlightened of the traditionalists simply force Linglish into those molds. Once the individual words are classified, traditional grammarians list and define, again semantically, the ways in which word classes relate to one another in phrases and clauses: Noun-Verb (boysplay) is a subject-verb construction m which the subject is typically the "actor." Verb-Noun (. . . play ball) is often a verb-object construction in which the object is the "receiver" of the action. The largest unit of all, the sentence, к again semantically defined as a "complete thought." Sentences are then classified, again semantically, according to whether they ask questions, give commands, make statements, or express an optativ* (Oh, that he wouidstopf) Traditionalists emphasize semantic definitions so strongly for several reasons, but largely because of a common definition of language on which

THE MATURE OF LANGUAGE

25

they base their analysis: Language is a means to communicate ideas, a definition that makes meaning central to the description of a language. When a traditionalist sets himself to writing the grammar of a dead language, his first concern is to identify parts of speech in older texts according to the principles lie has inherited from many centuries of language study. Once he has identified those parts of speech, he seLs to arranging them in paradigmatic orders that will show what kinds of endings the parts of speech occurred with. His next concern is to describe how those parts of speech function in sentences in terms that will help him explain what those sentences mean. An area of major concern to the traditional historical grammarian, for example, would be the tense, mood and aspect system of the verb phrase, both because in languages like Latin and Greek it was communicated through a very highly developed inflectional system, and because the system itself communicated a very complex set of meanings. The traditional historical grammarian of the year 2500 would try to explain the grammatical meaning of Modern English verb phrases like He could have been found, lie will be working, He has been being questioned in terms of the state of the action, the time relative to the present, past, or future, in terms of whether the action is real or unreal Another of his tasks would be to explain clausal relationships in terms of their meanings: Because lie left would be a causal relationship; (hough he (eft would be concessive; after he left would be temporal; if he left would be conditional, (35,97,134,174) PROBE.EM 2.4: The traditional definition of a noun is a word that names a person, place, or thing: boy, country, rock, aggression, ambition, mind,force, spirit, meaning, love, peace., honesty, coincidence, luck. The traditional definition of a verb is ttiat it indicates action: run, jump, imiate, resemble, reflect, weigh, stand, haw. Comment on the usefulness of these definitions in a sentence like Coincidence resembles luck or Honesty has many aspects. PROBLEM 2.5: Л subject is frequently defined as either the " a c t o r " of the sentence, or that which the sentence is "about." How would these definitions apply LO the subjects (in bold face) of the following sentences? (I) My car drives easily. (2) The key opened the door. (3) The table hasuikey on it. (4) George received a wound. (5) Bill resembles hisfather. (6) № lacks intelligence. (7) Somebody said the world is going to end tomorrow. (8) The wall crawled with roaches the size of у ourfist,like (he one on your collar. 2,6: Rather than defining language a& a. way to communicate ideas, ^ we began with the following definition: Lanfiuage is a way of jjer forming actions with speech. (I) How would a grammar be organized if it were based on such an assumption? How would units of speech be defined? (2) Are there

2Л

HISTORY AND LAIVdUAOE

any other preliminary definitions of language that would lead to different ways of organizing and writing a grammar ?

Descriptive Grammars The descriptivists, who as we noted came out of an earlier anthropological tradition, rejected the traditionalists and their use of meaning to describe the grammatical structure of a Language, first because the temper of the twenties and thirties stressed scientific objectivity, analysis of data without presuppositions, and a materialist, empirical approach in psychology that considered only objcctiuable, quantifiable, observable behavior as legitimate data. For behaviorist psychologists and their linguist colleagues, the mind as an observable entity did not empirically exist. Hence theories built around its supposed operations were invalid. This intellectual atmosphere, was reinforced when linguists found that many of the exotic languages they were studying required ways to categorize parts of speech and sentence functions that were quite different from the traditional ways. In some languages, the distinction between nouns and verbs seemed scarcely to exist. Nouns were discovered to be inflected for tense; the traditional categories of prepositions, articles, and adverbs were often impossible to apply consistently. The notion of verbs inflected for past or present was irrelevant for a language whose primary interest in verb inflection was not time but whether the speaker knew from his own experience that an event had occurred, whether it had been reported to him, whether it was an accepted fact. The traditional categories of perfect and progressive were insufficient for a language that inflected its verbs for the physical shape of their objects. The descriptivists, therefore, adopted the stance of the behaviorist psychologists and tricti to devise purely objective, entirely formal ways to discover and describe the structure of any language including Lnglish. First, instead of prescribing what English sentences should be like on the basis of their preconceived notions about sentences in Indo-European languages, they collected vast amounts of linguistic data, examples of sentences from letters, newspapers, telephone conversations, and so on. After identifying individual words and parts of words on the basis of how they were pronounced, they grouped the words into parts of speech primarily by what kinds of inflections they occurred before or after, and by their typical position in a sentence relative to other parts of speech. In English, for example, anything that occurs before a possessive or plural marker is (in one grammar) called a Class I word, noun being rejected because it is too closely associated with the traditionalists. Any word before a third person -s marker, a past tense marker, and an -ing ending is a Class I I word, verb being rejected for the same reason noun was. Descriptivists also

ТНК NATUItP. OF LANGUAGE

27

use position as a way to define parts of speech. Thus any word that occurs between ihe and a form of be is also a noun or Class I word: The f d . Anything that occurs after a Class Г word and ends the sentence is a verb or Class II word: The man This system generates several new categories of words, such as the category including jfj, no, uh-ftuh; the category including hello, goodbye, fit, so /с>Л£; the category including ivry, sotnewhat,ratker,quite. Once individual words arc put into these categories, descriptivists try to state their order relative to one another and how they group into phrases and clauses: veryaa6 old form a unit called an adjective phrase because the adjective ofd may occur after or without the very, but not vice versa: He was very old or He и-таг old, but not He was very. Larger relationships like subject-verb are again defined not semantically as " a c t o r " and "action" but formally: Whatever ihc verb agrees with (or might agree with) in number is the subject: The boy leaves vs. The boys leave. The largest structure, the sentence, is defined in various ways. Those descriptivists most strongly committed to the belief that sounds are the bedrock of language define a sentence as whatever occurs under a single intonation contour, a rising and falling of the voice that occurs between points of potential silence. Less rigorously phonological descriptivists define a sentence not :is a "complete thought" but as a complete structure composed of a subject and inflected verb plus all grammatical elements dependent on those elements. (59,66,84,2Щ Working this way seems perfectly suited to historical linguistics. Instead of collecting data from contemporary speakers, the linguist collects texts from an older period. He pores through the texts, discovering the pieces of words and whole words and how they fit together. He then groups the words into larger classes, into parts of speech, not on the basis of reconstructed meaning but on the basis of their inflections and positions. Then he tabulates how the various classes of words co-occur with one another to form larger units. He then describes the order of the larger elements. The complete descriptive grammar, historical or contemporary, consists of: I. 2* 3. 4.

An inventory of sounds and sound classes and how they occur relative to one another. An inventory of words and pieces of words put into various categories. Л description of how the pieces combine with the words or with one another, Л description of how categories of words combine into larger units and how those larger units combine with one another to form yet larger units.

2,7: (I) WhaL kinds of inflectional criteria would define a descriptive category we traditionally call adjective and adverb? {2) What problems result

28

HISTORY AMD LANGUAGE

from using only inflectional criteria with these words: beautiful, additional, representative, adjacent, tennis, ought, chaos, must, chief? PROBLEM 2,8: (1) Set up a category based purely on observable descriptive criteria that would unambiguously define the category of these words: may, will, can, shall. (2) Do the same for in, on, at, by, over, through, under, across, of, beside, between,behind, with. PROBLEM 2.9: What is the subject ofThere are men at the doori

Generative Grammars In the late fifties there occurred another revolution in linguistics. Intellectual revolution* begin when paradigms of questions exhaust themselves and someone recognizing that fact can formulate questions directed toward a quite different conception of и subject. This happened when MIT professor Noam Chomsky argued that the questions descriptivists had been asking depended on some serious misapprehensions they held about (.he naLure of language and the goals of language study. Chomsky argued that what we can discover about language strictly from collecting data made up of objectively observable units of speech will give us but a fraction of what we actually know about the mental grammar we have all developed from before the Lime we began speaking. Chomsky proposed that representing this grammar that we carry about in our beads should be the goal of linguists. This grammar is locked away in the brain, inaccessible to direct observation. But by asking what we know about our ability to produce and understand sentences, we can discover something of its properties and perhaps partially and very indirectly represent that ability in the grammar we write down on paper. (31, 32) Tn fact, this question—what indeed do we implicitly " k n o w " about sentences?—is the problem. We may not " k n o w " in the explicit sense that we can technically explain how sentences are put together. But we alt implicitly, tacitly " k n o w " a grammar. Otherwise, we could not speak our own language.

GRAMMATICALLY

Usage vs. Structure Obviously, the nature of the grammar we write down in order to represent the grammar we carry about in our heads is going to depend on what we discover we know about sentences. First, we know that some sentences are

THE NATURE OF LANGUAGE

29

correctly structured and others are not. The terms grammatical and шь grammatical are commonly used here. But they mean something quite different from junior-high-school ideas about "good grammar." We shall use these- terms as value-free labels to categorize sentences that do or do not correspond to the possibilities or" English sentences that are ordinarily used by any group of English speakers. We do not mean the prescriptions taught in junior high about not ending sentences with prepositions, not wing ain't, when to use shall instead of wilt, and s» on. These are problems of usage. Many teachers call them grammatical errors. And in a very special and restricted way. perhaps they could on occasion be called that. More accurately, though, they arc only grammatical patterns or word choices varying slightly from the patterns that some educated speakers use, often only m their selfconscious moments and usually only when they are writing. Indeed, most educated speakers, quite unselfconsciously, split their infinitives, end sentences with prepositions, use will for all persons and numbers. Unfortunately, except when past rhetoricians specifically commented on usage, it is almost impossible to discover much about the distant history of prestigious and non-preitigious speech patterns, A few comments about the appropriateness of the London dialect as a linguistic standard and the use of cenain dialects for comic purposes in Chaucer and elsewhere have come down to us from Middle English. But not until the sixteenth century do we find any specific observations about contemporary attitudes toward usage, and even then, the comments are often contradictory. It is a problem we will (shall ?) deal with again. (] 31) PROBLEM 2.10: Here are some sentences that deviate from standard English, some because they belong to a non-standard dialect, others because they violate a basic core of English structure. If you can identify those that would be used by non-standard speakers, how would your ability to do so serve as evidence thaL such sentences are in fact regularly structured with their own grammar? (I) He ain't can't have no money. (2) He ain't got no money. (3) He be have may gone. (4) He gone now. (5) She can may go. (6) She may can go. 0) Him and me donefinishednow. (8) They done may be finished with he and she. (9) Can't nobody tell him nothing. (iG) hnt somebody tell'him anything ?

A more significant kind of ungrammatically is that which violates the central core of grammatical structures shared by all speakers of a language. These next sentences, for example, would not be uttered by a native speaker °f English. Or, jf they were, he would cither correct himself, or, on reflection, i7je Lhat they were somehow Ll odd " : 1

• Street went down man the. * The man walked the do wn street.

2

39

HISTORY AMD LANGUAGE

3. 4.

The man wan! not down the street. Went the man do wn the street ?

All of them ate wrong for speakers of Modem English. But while the pattern of (1) and (2) would have been wrong at any time, (3) and (4) would have been correct for Shakes pea re. The second part of that conclusion is not difficult to arrive at since we find lines in Shakespeare's plays like 5. 6. 7. 8L

Know'st tkou Fluellen ?Непгу V. III.l. Stands Scotland where it did ? Macbeth. IV. j j i, 1know not bywhatpower/am made bold. MSN" D. Г ri, Nay, goe not from me.. . Ъ Henry VI. I.I.

But the first part or the cone [J si on has to be entirely inferential because we cannot ask any speaker of ModE what a speaker of OE or ME (Middle English)1 would think about such sentences as (1) and (2), We can only conclude that such sentences would probably have heen ungrammatical because we can find no sentences similar to them. The problem becomes much more difficult, however, when we wonder whether л sentence iike He should hate, been working would have been grammatical for, say, Chaucer, It is perfectly grammatical for a ModE speaker, but nowhere in ME do we find a combination Uksshetdd + have + been + V-ing. Does that mean it was ungrammatical and therefore never used ? Or does it mean iL just never happened to be written down? This is not a serious problem for the descriptivist or traditionalist who concerns himself primarily with describing the language found only in extant texts. Ft is a serious problem for a historical linguist if he looks upon evidence from the past as only a very small part of what a speaker might have written or said, and his intention is to create a grammar thai will account for more than just extant data. He must constantly ask himself whether structures that are possible for him in ModE but not attested to in OE or ME are missing from the texts accidentally or because they were not part of the grammar. 11 What might have been " is a crucial problem for such a linguist, (231) Ambiguity and Relatedoess Another ability that a modern transformationalist is concerned with is demonstrated in a sentence like The chicken is ready to eat and in the pair of sentences John is easy to please and To please John is easy. The first sentence is grammatically ambiguous. It can mean either that the thicken is about to eat or be eaten. The fact that any native speaker of English can recognize such 1 The dates assigned to O£ and ME яге arbitrary; most linguists put the end of OE at about AD. 1100, the end of ML al about л,п, 1500,

THE MATURE OF LANGUAUE

31

an ambiguity illustrates that we are able to recognize how the observable surface order of elements that make up the grammatical structure of what we hear or read can communicate very different underlying relationships among linguistic elements. This implies that a sentence has two levels of grammatical structure. There h the observable order of words and their observable connections, or the surface structure of a sentence. And there is a deeper structure that relates elements on a more abstract level In The chicken is ready to eat, nothing in the form of the sentence reveals that chicken can be either the subject or object of eat. But al some level of structure deeper than the simple order of elements in that sentence, we know that that is the case. Otherwise, we could not detect the ambiguity. M 2.11: All these sentences have two different meanings that depend on different grammatical deep structures. How are they ambiguous ? 1. 2<. 3. A* 5. 6.

The та» was too old to help. Flying planes can be dangerous. They called him a butler. He didis11 marry her because lie loved her, In conclusion, he said many foolish things. George was cooking in the kitchen.

We come to the same conclusion about grammatical surface structures and grammatical deep structures it' we ask how it is we know that the sentences, John is easy t ft please and To please John is easy are synonymous. They have very different surface structures, very different word orders, yet at a deeper level, they mean precisely the same thing; they have precisely the same set of abstract relations among their elements. In other words, related ness is the other side of the coin from ambiguity: Ambiguous sentences have one surface structure and at least two different deep structures. Related sentences have two surface structures and one deep structure. (94,132) And here again, if we take as one of the goals of our grammar a description of this fact, we raise some very difficult problems for the historian of language. Because these relationships can become very abstract and very difficult to describe even for a speaker of a modern language, they are even more difficult to describe for languages which no one any longer speaks. The point of this brief summary of the kinds of concerns linguists have when they ask questions about the grammar of a language is that as new theories of language come to dominate the thinking of those writing grammars p | contemporary languages, historians of languages are forced to ask new Kinds of questions about the data they are examining, indeed, to " s e e " data * a t linguists of other persuasions do not. As their understanding of what a lv 'ng language is like changes, so do their goals in describing a language no J ongerspoken.

32

HISTORY AND T.iNfi LI AC i H

lUit perhaps the most crucial point is that modem linguists no longer look primarily to the past to he!p them understand the grammar ol" a language, but into their own intuitions as native speakers of a human language. In other words, we will understand our linguistic history only when we understand our linguistic present.

LANGUAGE AND MKANING

In addition to the sounds and structures of the past, of course, we are concerned with words, their meanings, and the meanings of the sentences they occur in. But we have only to think how inadequate any dictionary definition is for words like lore, peace, or democracy to realize that most of what we call meaning cannot be written down. And when those words are combined into sentences like Peace with honor is justice in a democracy, we recognize that even if we could define individual words, those words in grammatical sir uclures add up to a total meaning thai is not the mere sum of the individual meanings. Traditional grammarians have always been concerned with meaning, hut meaning of a rather narrow kind; verb tenses, clausal relationships, kinds of nouns (common and proper, abstract and concrete), and so on. Descriptive grammarians set aside the problem of meaning in order to concentrate on observable forms and how they occur with one another. They talk in terms of grammatical meaning that can be abstracted from lexical meaning, of the grammatical meaning o£'T«esbriSig, ami tbcsShhy tores / Bid gyre and f>imt>k> in the wahe. . . . More recently, however, transformational linguists have turned baek tn lexical mea uing because they have set as their task a description of how semantic content is " ma pped " o n t o syntactic and phonological forms. Unfortunately, no area of language study is more obscure, more difficult to formulate in terms that would let us study it- The meaning of "meaning" has been debated at least since the pn>Socralic philosphers of ancient Greece. It is a question that has so preoccupied modern thought that some philosophers have claimed that philosophical problems arc really linguistic problems and that the proper task of philosophy is to untangle linguistic со п fu si о n s i n t h e way we pose ou г q ucstions r Because different disciplines have approached the problem of meaning differently, it is no surprise that there are as many different ways to describe meaning as there ane disciplines. Neurologists define meaning as the selective activation of neural networks at varying amplitudes and frequencies in response to linguistic and nonlinguistit stimuli, resulting in chemical changes in the molecular structure of large molecules. Psychologists have defined meaning variously as mental images, as conditioned substitutions for word referents, as dispositions to behave, as the network of associations a word is

THE NATURE ОГ LANGUAGE

33

part of. Philosophers have defined meaning in a multitude of ways: as how a word or sentence is used, in what is necessary to confirm its truth, in the existence of wo rd~refcrents, in the "essence" of a referent. . . . (191,207) When a question about the meaning of a partkubr word arises, though, we usually think first of the dictionary, where along with a good deal of additional information about spelling, pronunciation, etymology, and so on, we can find its ''meaning. " Л typical entry minus all irrclevancies might look like this: Chair: A scat for one person; now usually the four-kgged scat with a rest for the back. Typically, the definition puts the word to he defined into a larger or suptrordinale class, in this case seat. Seal names the larger class of concepts to which the concepts covered by chair belong, along with the concepts covered by stool, bench, hassock, settee, perch, and so on. Once the superordinate category, seat, is named, those features which distinguish the word chair from those other words otganize the definition into finer and finer subordinate classes: Seat (to distinguish from perch) (to distinguish from ioie seat) (to distinguish from hassock)

for back

(to distinguish from stool)

We may choose different kinds of subset differentia to distinguish members of the superordinate set, choosing criteria by what the object is used for, as we would have to with hammer; how it is made, for net; what it is made of, for glacier; what it does, for heart, and so on. But if the word is riot defined merely by listing synonyms or pointing to a picture, the pattern °f definition is invariably the same: The word to be defined is placed in a superordinate set and then distinguished hy subordinate differentia. A problem of theory in regard to dictionary definitions is that every definition depends on words the reader already knows and the words that ha to be in the language. This, of course, finally leads to theoretical

34

HISTORY AMD

circularity since the words used in a definition must themselves be defined by other words, which in turn must be defined, and so on. PROBLEM 2.12; Pick any fairly general word in the dictionary that would be used in the definition of a more specific word, a word like shape, substance, man, good, and look up a key word in its definition. Then look up the key words in its definition. At how many removes do the definitions become circular, referring back to an earlier word ? Recently, linguists have approached the problem of meaning from a different direction. When we examine the meaning of groups of words that belong to the same semantic space or lexical field, we often find they occur in symmetrical sets. Kinship terms, for example, always occur in sets whose members are distinguished by only a few component of meaning. The Only si: man Lie distinction between the members of the pairs husband-wife, uutituncle, sister-brother, son-daughter, father-mother is whether the referent is male or female; father-brother-sort, mother-sisier-daughJer, auttt-cousinniece, unde-cousin-nepnew differ only in whether the referent is in the ascending, descending, or the same generation. Ssbling-cauxin, nephew-son, parentUftdejauni differ in whether the referent is a direct lineal relation or once removed. In fact, we could define all these words more economically than a dictionary would by specifying only which of the pairs of complementary semantic features are in the meaning: 1. 2. 3. 4. 5. 6. 1, 8,

father: [ + male, + 1 generation, + lineal) mother: [—male, + ] generation, + lineal] son: [+male, — 1 generation, + lineal] daughter: [-male, - I generation, + lineal] uncle: [ + male, + 1 generation, —lineal] aunt: [—male, + 1 generation, —lineal] brother; [+male, 0 generation, -f lineal] sister:[ ~ male, 0 genera tion, + li iieal](J2,203)

2.13: (I) What features compose the meanings of nephew, niece, cousin, parent, relative, father-in-law'1. (2) Are there clusters of semanticfeatures for which we have no words? What word means [± male, —1 generation, -lineal]; for [-male,0generation, -lineal]? PROBLEM

Originally, these techniques were of interest only to anthropologists interested in describing tajionomic semantic systems such as kinship, disease, and plant names. More recently, generative semanticists have attempted to adapt such a system to describe words in general. The meanings of idea and rack, for example, contrast in one basic way: abstract vs. concrete, a contrast we can represent as [+ concrete]; between grass and snail is the opposition

THE NATURE OF LANGUAGE

35

[+animate]; between snai/and woman, [± human]; between woman and girl, [± adQIl]; between girl and bvy\ the familiar contrast of [±male]. (29, 101, 102) Because this approach to meaning is still in Us formative stage, many problems remain to be solved: the kinds of conceptual opposition*, the number of features, whether some features might require a scale of numbers rather than just [I ,.'•--]. Most importantly, it is clear that the features must be grouped and structured. The definition of chair as a seat with a back for one person to sit on would have to include at least this information: person

with legs J

four In other words, we are dealing with meaning as if '*a meaning" were decomposable into molecules, atoms, and perhaps even atomic particles, all of which are organized into very complex structures. When this theory of meaning is applied to problems of historical semantics, however, the same problems of describing grammatical structure in dead languages arise, but more severely. For this theory of meaning, like the theory of generative grammar we described, relies on introspection, on asking oneself about meaning and contrasts of meaning. But this is no simple task even when we ask about our own language. It is very difficult to recover this kind of meaning entirely from the contexts in which we find words in those relatively few tests a dead language has left us. What this also demonstrates is the futility of looking Lo older forms of a language to tell us about our own. We know infinitely more about the meanings of words in the language we speak than we ever will about words in any dead language. So again, we cannot begm with the past in constructing a theory of meaning, much less a theory of language. To summarize the point of this very superficial outline of some theories about linguistic structure: When we ask questions about the nature of language or about its history, the theory we base our questions on will determine •what questions we ask. The assumptions we base our theory on whether we look to the past for a more perfect form of grammar and consequently have to rely on writing as the source of our information, or whether we rely first on what we can discover about our own linguistic competence, as difficult as it

36

HISTORY AMD

may be to probe and represent in a written grammar—such assumptions will strongly determine the kind of theory we construct.

SOME CAUSES AND EFFECTS OF LANGUAGE CHANGE-: Now these theories will provide us with a way to describe what has changed in the internal structure of our language, Hut a grammar (including phonology and semantics) is only one component in what we might term a theory of language- Language is more than its grammar. It is an instrument by which we perform most of the social acts that constitute a society. It reflects, indeed may on occasion determine, our social class. Some believe that it even determines how we structure our perceptions of reality. In short, language is part of a larger set of interlocking cultural and social structures that we move in from the moment the doctor announces It*s a , with alE the social values the terms boy and girl entail, to the obligatory final ritual (most stereotypically expressed in westerns and nautical movies, when someone is required to "say a few words" over a body). We can recognize how highlyvalued language is in our culture when we celebrate the first word a child utters and make collections of the last words of great men. In such a contest, language will change when culture and social conditions change. And it is in these external changes that we may be able to find some of the causes of internal linguistic change. We have already touched on one obvious source of linguistic change: external social forces. The invasion of England by the Danes and Normans profoundly affected our language. So did printing. So did the social turmoil of the fourteenth century. So has our science and technology. Another important social source of linguistic change are the differences among geographical and social dialects. Inevitably, given several social/geographical dialects, a prestige dialect will emerge to influence and be influenced by other dialects. Tn order to explore this aspect of language, we must describe the social conditions in which language is used and the elements that may cause a rigid social structure to become fluid, so that speakers from different classes wilt come into contact with one another. But other sources of change may be found in the internal structure of the grammar itself. This will be a very difficult question t" discuss because we have to treat the structure of the grammar as if it were a real model of the speaker's mind. Once a native speaker unconsciously formulates in his mind the grammar of his language, that grammar may be very well organized and very well integrated. But at certain points, it may also be unnecessarily complex or redundant. The simplest example of this kind of inefficient complexity in English is found in our verbs. The most economical rule that would

THE NATURE OF LANGUAGE

37

account for the past tense of verbs would be something like this: Verb -Everb-ed. Jn ModE, though, we have to list well over a hundred exceptions to this rule: buy-boughs,sellsoM, siftgsmg, and so on. But in OE, there were hundreds of exceptions. We seem to be very slowly simplifying a point of unnecessary complexity ill the grammar. The reason speakers simplify their grammars is undoubtedly to be found in some- kind of principle of least effort that all humans rely on when they construct schema or models for their behavior and understanding But this principle probably operates most powerfully among children as they learn a language. Children, for example, often go through four stages when they learn irregular verbs. In the first stage, they often are able to use irregular verbs correctly: He saw. In the second and third stages, they combine the irregular past or the infinitive with the -ed ending: He sawed and He seed. Finally, they learn to use the verbs as they are used in the dialect they grow up in. If it happens to be standard English, then they will eventually learn to say Ш saw again. But between every generation there is probably a slight grammatical discontinuity. The language is not passed on from parent to child perfectly. The child will make some mistakes. If enough similar mistakes are made by enough children, then those mistakes may become part of the grammar tor that generation of children, a grammar that wili be passed on to the next generation. These, then, are the kinds of problems we shall investigate. Because we have to understand the social and cultural contexts in which changes occur, we shall begin by examining those contexts through the sources of our vocabulary, an aspect of language particularly sensitive to cultural changeThen in Chapter Six we shall deal with how we create words out of the basic material of our lexicon, and in Chapter Seven how the meanings associated with those forms evolve. In Chapters Nine through Fourteen, we shall examine how the grammar and sound patterns of our language have changed. PROBLEM 2,14: (1) What are some occasions in our culture, other than birth and death, where the use of language is essential, where if certain words are not spoken, the moment loses its significance or becomes entirely invalid? (2) Why does the Gospel according to John begin as it does? (3) What would it feei like, do you suppose, to be an adult and literally have no name? Not to have forgotten it or to be unable to lind it, but literally to have been given no name ever? (4) Are there any social situations or events where speech is tacitly forbidden? PROBLEM 2Л5: Between the bee and man there is a great linguistic gap. In a series of minimal steps, evolve the language of bees until it approximates the communicative capacity of man. Bees communicate distance by dancing increasingly less intensely as distance from the hive to the source of food increases. Direction is indicated by translating the axis of the sun-hive to a

38

HISTORY AND LAJ^GLWGE

vertical line inside- the hive and the» by dancing at an angle from vertical corresponding to the angle formed between the sun-hive-food source, Tlieir mode of communication, of course, would have to develop out of the intensity of their buzzing and the movements of their dance. PROBLEM 2,16: What would be the minimal unequivocal evidence that a primate now being trained to communicate via sign language or plastic counters had the ability to communicate on a ievel approximating human communication? (It is almost certain that the mode would not be verbal, since the Jaryngeal structure of primates is not well adapted to the production of finely articulated sounds.) PROBLEM 2,17: The boldface examples in tlie following sets are considered "ungraminatical." Construct a logical argument that would support the claim that they are preferable to their grammatical counterparts, 1. 2. 3. 4.

Myself, yourself hisself, herself, ourselves, Iheirscl ves. I am 't, у аи aren' ?, it isn't, we aren't, (hey aren't. I wasn*r, you wasn't, it waSn t, we weren't,you weren't, they weren't. I jumped, you danced, she ruiuicd, we singed, you gfted, they worked.

PROBLEM 2 J 8 : Here are several words from the semantic field referring to the use of language. They all seem to imply judgment, (1) What can you conclude from these words about our attitudes toward how people use language? (2) Are attitudes toward the use of language the same in all parts of the English-speaking world? Are there stereotyped ethnic or sex differences in language use? (3) What use of language appears to be most highly valued, reticence or поп-reticence ? 1.

talk, chatter, gab, prattle, gabble, babbie, jabber, clack, gossip, chat,

2.

talkative, gabby, wordy, gossipy, soluble, loquacious, garrulous, prolix, diffuse, bigmouthed, fafmouthed, glib. shrew, nag, gossip, fishwife, chatterbox, windbag, higmouth, fatmauth, magpie, scold, harpy, termagent. taciturn, Income, reticent, concise, trenchant, brief, terse, pointed, pithy, succinct, crisp, sententious, epigrammatic, elliptical, crabbed, curt.

3. 4.

PART II

WORDS AND MEANINGS

Chapter 3 WORDS: NATIVE AND BORROWED

OLD ENGU5EI AND В КРОНЫ Of ah the aspects of language that charge through time—pronunciation, spelling, usage, grammar, meaning, and vocabulary—meaning and vocabulary are the most sensitive to the external social and historical forces that determine which words a culture preserves from its own heritage and which words it borrows from others. Particular meanings change themostidiosyncratically. We could never have predicted that the meaning once attached to dwelt would shift from " lead into error" to its current sense. There are some large generalizations we can make about how meanings change, but they are relatively few. The total lexicon of a language, however, is a very accurate linguistic barometer to the broad social and historical changes in the history of a culture. Given the relevant cultural information, we could have predicted fairly accurately the general makeup of our ModE vocabulary and its sources. We know that some kinds of words arc so durable that they will for centuries r esist the most violent vicissitudes of time: hand, foot, mother, father, head, heart, sun, moon, sleep, eat, and so on. Other words predictably disappear: OE wergild, the money a murderer had to pay to the family of his victims; 41

42

WORDS AND

OE к и т , a hairy garment; ОС freox, a hurting spear. Cultural necessity has forced other words into the language through borrowing, derivation, or creation: machine» cab, crime, jazz, etiquette, chocolate\ albino, ghetto, syrup, candy, shawl, jungle, iycoon> taboo, tote, blender, television, retrorocket. The Origin nf Words How and where the first words originated has always puzzled linguists and philosophers. When Socrates and Hermogcncs debated the question 2000 years ago, they argued whether words related naturally to the things they named. It is it notion that helps explain much of the magic and folk belief found almost everywhere in the world. The Egyptians, for example, gave everyone two names, a public name and a secret one; they believed if someone knew a person's real name, he would have power over the person. In other cultures, word-magic takes the form of taboos against uttering the name of a god, or the name of certain relatives, or even words that sound like those words. (60) The relationship between a word and its referent is for the most part arbitrary. The shape or sound of a word has no natural relationship to the thing it names. There may be a narrow range of words that could illustrate phonetic imitation: boom, clangs hiss, screech, and so on. And a few others may illustrate principles of conventionalized phonetic symbolism: Teeny is smaller than tiny, for example. But among the hundreds of thousands of English words, they are a very tiny (even teeny) minority. So if primitive man did create words by imitating sounds around him, the words have changed so greatly that almost all traces of the original association have disappeared. The sound of the word dog does not resemble one. The other speculated sources are equally problematical: the grunts and groans of work, cries of joy and sorrow, sounds made by the tongue imitating the shape of an object. Nor does it help to study the language of the most primitive peoples in the jungles of South America or the Philippines, Such languages are as mature, as complex, as expressive for the needs of their speakers as any modem European, African, or Asian language. Thus what we must begin with as we seek out the history of any specific language are not speculations about fancied linguistic prehistory, but the "givens," the oldest words that we can trace back to the earnest texts. Oldest Words in Knglish In English, the oldest data are the words we can find in written texts and carved monuments surviving from the Old English period (c. A.D. 450-e, 1100). 3.1: While it is not easy to explain why every item among the words left to us from OE should have endured, been replaced or lost, we can make

WORDS Г NATIVE AND RORROWEn

43

some general preliminary observations if we examine this Ikt of words. Those starred have beer completely lest from the language. The others are direct ancestors of the Modern English word found in parentheses. Why have we lost some and retained others? * coward (consecrated bread used as a lest for truth), "dolgbot (compensation for wounding), wtf (wife), jod (food), *pqft (bench for rowers), *scofa (a hairy garment), stars (stone), *peox (hunting spear), winter (winter), *eafor (tenant obligation to king to convey goods), god (good), *fiytme(& hiood-letting instrument), wster(water), *feohfang(the offence of bribe-taking), brodor (brother). *eam (mother's brother), *nrsed (quick), *barda (beaked ship), com (corn), blod (blood), hand (hand), gntnd (ground), kind (land), *faSe (father's sister), win (wine), heorte (heart), keo/od (head), lufit (love), siepan (sleep), *steting (hunting rights), silian (sit), *zwitl(a narrow-necked basket), > A Note on Pronunciation: The letters in OE had roughly the following values. <5> and <в> were pronounced like in thing when they occur at the beginning or end of a syllable or next to a

, , , or . Otherwise they are pronounced like in the, and <s> were pronounced like and <s> in fit and sit at the beginning or end of a syllable or next to a

, , , or . Otherwise they were- pronounced like and respectively. <sc> were pronounced like <sh) in ship before Lhe Setters and <e), like <sk) in skip elsewhere. At the end of a word or before a consonant (h> had the German quality of in Back. Before or after or <e>, was like a heavily aspirated in yield; before or after back vowels , , it was pronounced like the voiced equivalent of German was always pronounced after , as in ModE fonger. OF. vowels were either long or short, but since OE manuscripts did not indicate quantity by diacritical marks or spelling, we have not used length marks here. The vowels have their continental values: as in see or sit; <e> as in bate or bet; <ж> as in bar от a lengthened pronunciation of bad; as in hot or a lengthened pronunciation of hod; as in bought or boat; <4> as in рш от pool. In early OE, was like a long or short German .

THE MOST DISTANT ORIGINS: INDO-FCUROPEAN If we can infer a good deal about an older culture from the words it no longer uses, we can also discover a good deal from the words it passes on. From the w d in Problem 3.] and from others, we know that those which have been

44

WORDS AND MEANINGS

preserved cover some of the most basic objects, actions, and concepts of daily Jife, words like hand, food, wife, sun, hause, stone, go, sing, eat, see, sleep, good, wise, cold, sharp, in, on, off, ot>er. These concepts are so independent of particular cultures, so basic to human life that it is almost certain we would find in ail languages tbaL words for these concepts have been passed on from generation to generation for centuries, pronounced and spelled differently, perhaps, but basically the " same" word. (3,44,128) PROBLEM 3.2: Words from several languages that refer to roughly the same concepts are shown in Table 3.1. What do you conclude from the fact that in some cases, among several languages, roughly the same mean ing is represented by words that are rather similar to one another, but in other cases are notV That is, night is rather close to Sanskrit nakiam but very different from Japanese ban. FROBI-FM 3.3: Here are some words in various languages for aluminum: French: alamtttum, Spanish: atuminio* Italian: вНшптю, Dutch: aluminium, Danish: aluminium. Polish: aluminjum, Hungarian: aluminium, Turkish: аШттуот, Indonesian: aluminium, Russian: alyununi, Arabic: akminyoum, Japanese: aruminyuumu. Why are they alike ? As Problem 3.3 demonstrates, words can resemble one another from language to language because they have been borrowed from some common source. Hut when we consider the likelihood of borrowing the word for aluminum and the likelihood of borrowing words so basic and common as mow, night, hundred, and so on, we can also tentatively reject borrowing as an explanation of widespread similarities among the most common words in different languages. The more plausible explanation assumes that in each language, the words must have been inherited from some common ancestor language, and that through time, in different descendant languages, the forms of the words gradually changed. Once we establish the principle that similar words with similar meanings (or meanings which at one time we might speculate were similar) may be descended from some common but now lost ancestor form, it becomes possible to reconstruct in very rough outline some of those earlier ancestral words. If, for example, we compare the word for mother in the languages we suspect are related to a single ancestor, we can create a form from which the recorded ancient and modem words for mother can be consistently derived. Compare these words: English mother, Dutch tnoeder, Icelandic mofyr, Danish moder, Irish mdthir, Russian mate, Lithuanian mote, Latin mater, Persian madur, Sanskrit matt: From the features these share, we could postulate as the parent form this hypothetical root: * mater. Lach letter in the root is a symbol from which we can derive by means of a set of phonological rules

COGNATE AND NON-COGNATE WORDS

English German Dutch Swedish Latin French Spanish Italian Rumanian Greek Polish Czech Russian Sanskrit Hungarian Finnish Turkish Arabic Swahili Japanese Chinese

night nacht nacht natt noctis nuiL nouhc notte noapte nuktos noc noc

noch nakta ejszaka yo gcoe lay la usiku

ban wan

snow schnee sneeuw

seven siebtn zeven

foot fuss voet

sno

sju

fot

nlvis neige nieve neve zapada nipha snieg snih snye'k snehai

septein

pcdis pied

ho

lumi kar gal id thcluji yuki hsueh

sepc

sielft scltc sapte hepta siedem sedm sycm

pie

sapfca het

pal lab

seitseman yedi sabaa saba shichi

jalka ayak

chi

piede picior pod os stopa no ha naga

qadam niguu ashi chiao

fish usch vis iisk piscis poisson pescado ренсе peste psari ryba ryba riba tnatsyah hal

kala balik samak samaki sakana yii

heart herz hart hjarta uordis CftUT

corazon cuore inima kardia serce

sidce sye'rtse hftlsziv sydan kalb qalb moyo shin hsin

hundred hundert honderd hundra cenlum cent dcnto cento suta hekaton Mo sto

sto satam szaz

sata

yuz maah niia hyaku pai

ten zehn ticn tio

decem di?t

diez dieci 2£СЙ

tooth zahn tand tand dcntis dent diente dente dinte odontos

cfeka dziesitji deset dyiisit dasa

zub zup dant

tiz

fog

kymmenen on aiihara kuma juu shih

Tab

liammaa dis fin

jino ha che

46

WORDS AND MEANINOS

the sounds of cognate words found in dcsccndanL languages. It does not necessarily represent the way the ancestor word was actually pronounced at any given moment in our linguistic prehistory, though it very likely is reasonably close to it. Certainly, Indii-European, the name of the reconstructed hypothetical common ancestor Language, was itself once a dialect or collection of dialects of some even more distant progenitor. Some linguists have attempted—and failed—to group them with the Hamito-Semitic languages (including Arabic, Hebrew, Aramaic, Coptic, Berber, and the North African Cushitic dialects) or the Finno-Ugric (Finnish and Hungarian). But no one has found enough evidence to confidently relate the large and scattered group of modern Indo-European languages with any other language family. PROBLEM 3.4; We have seen that from OE words and their meanings, we cari deduce something about Anglo-Saxon culture, even if we had no firsthand knowledge of England, its location, or its climate. OE words for referents like the ocean, winter, ships, deer, fish, oak trees, chalk, and so on would lead us to a Northern European location somewhere close to the sea. Numerous words for concepts in law suggest an elaborate legal code based on duty <snd payments. Words for mother's brother and father's brother suggest a kinship system more complex than ours and one that seems to emphasise male kinship structures. Here are some data (some of it misleading) about words common to Indo-European languages, plus some geographical, botanical, and sociological data that wiil allow a rough guess about the general area of the original Jndo-European homeland. (1) Sanskrit, the oldest of the IE languages with extensive extant documents (с. 150Пв.с.) was spoken in Northern India. (2) Tobacco, referring to a plant now found around the eastern end of the Mediterranean, is found in almost all modern IE languages. 0) Cognates for the following wfinis or other words for their referents are found in a wide variety of IH languages: snow, freezing cold, winter, summer, spring: oak, beech, birch, willow; bear, wolf, otter, beaver, weasel, deer, rabbit, mouse, ox, horse, sheep, goat, pig, dog, snake, tortoise, ant, eagle, hawk, owl, herd, salmon, cow, udder; cheese, mead (a fermented drink containing honey); wheel,
WORDS: NATIVE AND EJOKROWF.D

47

of Asia. (7) The salmon is found in northern European waters and a similar fish is found in the Caspian Sea. (1) What can be reconstructed of the culture of those IE speakers? (2) How might we estimate the approximate age of IE from cognate words? Exactly what happened five or six thousand years ago is, of course, impossible to reconstruct. But it is likely that for some reason, groups among the IE tribes began migrating first to the east and south, then in all directions from their original homeland. Not long after, their language, probably already more a collection of dialects than a single uniform tongue, began to change until the dialects became mutually unintelligible languages. With no written standard and wiLh virtually no significant contact over what for their speakers must have been immensely long distances, nothing interfered with the natural tendency of every language to change. (7, &, 23,62,2 IS)

Iridu-Iluroiiean > West I uropcan

It has been thought that IE first split into Eastern and Western branches because of the widespread correspondences of one particular sound change east and west of a line running roughly north and south at about 2011 east latitude. East of this line, the original *k- sound in IE changed to a sibilant, a s or sh sound. The IE root for hundred, *kmldm, became sat am in Sanskrit, Simtas in Lithuanian, suto in Old Slavic. In the Western branch, it remained k, as in Latin ceirfum and Celtic can/, then changed to ft in the Germanic languages : hundred, or to s or ch in Romance languages: cent, cie 3.5: Does this confirm or contradict your conclusions about the tE homeland? Why? The Eastern branch then "split into two: (I) the Ralto-Slavic, which includes Lettish, Lithuanian, and Old Prussian among the Baltic; and Bulgarian, Slovenian. Serbo-Croatian, Polish, Czech, and Russian among the Slavic; and (2) the In do-Iranian, which includes modem Persian, Hindi, Bengali, and Romany—the traditional language of the Gypsies (a word adapted from Egyptian, from whence Ihe Europeans believed them to have come). The Western branches split into aL least four more branches: Hellenic, Italic, Celtic and Germanic Most scholars also include a dead language discovered in the early years of this century: Tuebarian, surprisingly found in Central Asia, far to the east of the Western IE languages, which it resembles •n some important ways. It was probably spoken by a group that originally

4S

WOKDS AMD MEANINGS

belonged to the Western branch but shortly after the Centum-Satem split (as it has bctn called), migrated eastward. One other language, Hitiite, evidence for which has been discovered in Turkey, is also included among the IE languages, though il is unclear exactly how it related to the two main branches. PHOELEM 3.6: Here are a few cognates in the Western branch that do net appear in the Eastern. Comment, corn, groin, furrow, bean, meal, maw, sea, salt, fish, elm, finch, starling, swallow. Of the several Hellenic dialects, Attic Greek, spoken in Athens, became the standard, a natural consequence of its being the political and cultural center of the early Western world. From the Italic descended two dead languages, Oscan and Umbrian, and Latin, from which descended French, Spanish, Italian, Portuguese, and Rumanian. Celtic split into the extinct Gallic, Gaelic (the ancestor of Manx, Scots Gaelic, and Irish Gaelic), Britannic (the ancestor of the now dead Cornish and Pictish), the dying Breton, and the mildly robust Welsh,

> West European > Germanic > West Germanic The most important subgrouping for our purposes is Germanic. Its earliest records go back to some fourth-century Scandinavian inscriptions and a translation (by Bishop Ulfilas [с, 311-3&1]) of parts of the Bible into Gothic, a now extinct East Germanic language. The largest body of early literature appears in OE after A D . 700, and in Old Icelandic after 1100. Germanic is conventionally divided into three branches on the basis of certain phonological and grammatical changes that occurred before about A.D. MM): (]) East Germanic, which includes the dead Gothic; (2) North Germanic, which includes two groups: (a)Icelandic, Norwegian,and Faeroese (from the Faeroesean Islands); and (b) Danish and Swedish; and finally (3) West Germanic. This includes Dutch, Flemish, Afrikaans, Low German, modern standard German, Yiddish, Frisian, and English. PROBLEM 3.7: Cognates of these words are found only in the Germanic languages. Comment, broth,, brew, dough, knead, loaf, wheat, gold, silver, lead, tin, buy, ware, worth; borough, king, earl; book, lore, write, teech (healing): cliff* island, sea, sound (as in Puget Sound), strand (beach); whale, sea! (the animal); ship, steer, sail, north, east, south, west.

K: NATIVE AND BORROWED

49

PROBLEM 3.8: (I) What is dangerous about relying on negative evidence in attempts to reconstruct cultures or geographical origins from linguistic delta? That is, what does it prove when a number of languages known to have descended from the same ancestor language are shown not to share cognate words for fish! What does it indicate when we discover that cognate words from the root for hand are found only in Germanic languages? (2) What further problem in cultural reconstruction does the following example introduce? In Great Britain, the word robin denotes a red-breasted member of the warbler family. When the colonists arrived in North America, they found a red-breasted member of the thrush family. They called it robin. 3.9: We can show how the Indo-European languages relate to one another by means of a tree, as shown in Figure 3.1. This figure is a model of INDO-EUROPEAN I -ЛЧ O-

WESTERN

, CELTIC

I

г

GFft MANIC

ITALIC

NORIH

WEST

•

HLLLENiC

TOCHARIAN

НЛ1 ГО-SLAVIC

INDO.UMNIAN

HITTITE LAS I

Figure 3.1. Relationship of indc^Huropean Languages the historical relationships as well as the linguistic relationships. How does it lead us to think about the way one language splits into two or more languages? What problem does the following diagram and explanation introduce? (14, 171,204) Western Celtic

Eastern

Germanic *

Italic

4

* Hellenic

Balto-Slavic

I ndо-Iranian

Both the Germanic and Balto-Slavic languages have a similar inflectional ending for instrumental plurals. Other languages have endings related to a different sound. The Celtic and Italic groups have a similar passive voice inflection. The H elleni с а п d I n do- T га ilia n have simila r past tenses. The Hellenic- and Italic share a. characteristic of feminine nouns with masculine suffixes. ?• The Germanic and Italic use the perfect tense as a general past lense.

50

WORDS AND MEANINGS

From the common vocabulary, archenlogical remains, and the observations of Roman historians, we can sketch the outlines of pre-h is lu He Germanic society. Because their common vocabulary included for the first time many words referring to advanced agriculture, farming must have become more important than it had been. More significantly, the ocean had also become important, ft would be the Viking long-ships that would carry the Germanic warriors across the seas to raid, plunder, and conquer from Britain to France to the Mediterranean. Their social and economic organization must also have begun to develop. King, earl, and borough indicate a government and an incipient feudal society; gold, silver, lead, tin, buy., ware, and worth indicate an economic life beyond trading in kind. They were a diverse lot, though. They included the Franks, the Goths, the Vandals, and the Lombards, all warlike enough to harass France, Spain, Rome, and Africa and give the Teutons their fierce reputation among the Roman historians. They also included the- Germans, who did not wander far from Central Germany, and the Northmen (hence- Norseman, which finally became ffvrHlOn), who both farmed and sailed. They shared a common mythology of Odin and Thor and an epic poetry chat celebrated the values of honor, loyalty to chief and kinsman in return for their generosity with gifts, and bravery and gtory in battle. (103) From certain La Lin words borrowed into Germanic before Lhe AngloSaxons invaded Britain in the fifth century, we know that they must have had some contact with Rome, (7,194) PROBLEM 3.10: From these borrowed words, speculate about the kind of contact the Germanic tribes had with Rome. The first word in the list is the original Latin word. The second is the OE adapted from Latin with a modern translation in parentheses if the word has been lost. The third is the descendant ModE word or its closest equivalent. Where the word has been lost in Mod L, the symbol 0 appears. camputti-ettmp (field, battle)-^ tribirtum-infei (tribute)-P mango-mangim (to barter with)-monger (asmfishmonger) Шопёит-toH-iott • poudn-pund- -pound tttitiupassutn (a thousand steps)-m:I-mile ralcem-ceatc-chatk cuprum—ccpor—copper pic- -рк-pitch {the substance) busyrum-buiere-butter ulmim— win—wine

WORDS; NATIVE AND BORROWED

SI

merttha-minle-mmt (the plant) pisum-pisa-pea piper-pipvr-pepper prumtm-plume-plum plan ta-pfanie-pfant balteus-beh-belt soccus-socc-sock catillus-cite{-$ (ModE kmie borrowed il from Latin)

is borrowed from Danish, which also

plpa-pipe-pipe bsrtna-binf^-bin cuppa-cuppe-cup p&rma—pimne-pnn coquJtta-cycene-kitchen piima-pwn-pirt gi!mfw-gOTWi(gem)-0 (ModE gem is borrowed from French) ltnea-fitie-Une uatiuny-weali-wali

febris-fefer-fever

P«-Anglo-Saxon Britain Long before these northern Germanic tribes attacked the native Britons (orCelts), the Romans had long since raided, invaded, colonized, and deserted the island. Julius Caesar (100 B.C.-44 B.C.) invaded Britain twice, failing the first time in 55 B.C.; but the nest year with a larger force, he conquered the bland. Though he had invaded Britain to shore up his northern flank, he was also looking for slaves and tribute. Finding neither in sufficient quantity or quality to justify his effort, he turned from Britain to his problems in Gaul, giving the island a brief period of freed о in from Roman domination. Then in A.D. 43, Claudius (10 B.C.-A.D. 54) invaded ihe island, and after putting down an uprising led by the Celtic Queen Boadicea, finally brought Britain into the Empire. But because Rome was unwilling to expend the men a 4d effort to conquer the Picts in the wilds of Scotland while being harassed from the rear by the still unruly Southern Celts, its sway ended at Hadrian's Wall along the northern bank of the Tyne in the Lowlands. Thus, Roman cvviljzatiun was limited to what is now known as the Midlands and the Southeast, where Romans buiit their walled towns and villas and connecting roads in an attempt to reproduce a sunny Mediterranean life on (what was to

52

WORL>S AND

become after a global climatic change) a wet and cloudy distant outpost. (13, 34)

ANGLO-SAXON BRITAIN: THE BEGINNINGS It was to be shortlived. Before the end of the fourth century A.D., Scandinavian raiders from the north had already begin to harry the British coast. Simultaneously, the Picts
WORDS: NATIVE AND UORROWQD

S3

for peace and for the welfare of (heir native land and strive against their energies. And they gave them subslance and property for their struggle. They came from three peoples, the boldest of Germany, fro in the Saxons and the Angles and the Jutes. Concerning the Jutes, in the beginning they are in Kent, and the Isle of Wight; that is, the people who dwell in the Wight Island. From ilie Saxons, that is from (he land which people calf Old Saxony, come the East Saxons, and the South Saxons and the West Saxons. And from ihc Anglia come the East Angles and the Middle Angles and (he Mercians and all the Northumbrian people. The land between the Jutes and the Saxons is called the Angulus; it is said that from the time when liiey departed until today, it remains waste. So, it was at first thought that the Jutes came from what we now call Jutland; the Angles from the Western side of the Jutish peninsula and the east bank of the F.I he; Lhe Saxons from Lhe F.I he to perhaps the mouth of the Rhine. More recent archeological evidence locates the Angles farther southeast and the Jutes on the coast, near the Frisian Islands off the coast of Germany and the Netherlands (see Figure 3,2).

Figure 3.2. Origins of Invaders and Raiders

54

WORDS AND MEANTNGS

Figure 3.3» Old English Dialect Areas But then it is not entirely certain that the Jutes ever existed at all. Two other Old English commentators never even used the word Jute. They called all the invaders Saxons, or Angles and Frisians. Bede himself never used Jute again nor does it occur in Kent plute-names. A rcheo logical evidence, in fact, suggests that Kent was settled by a variety of groups. Indeed, the traditional idea that whole tribes moved en masse from the Continent to Britain may be wrong. It may be that small groups moved to Britain, fanning out from southeastern England to settle the west and north. The dialect areas that existed in Anglo-Saxon England (see Figure 3,3) may not reflect patterns of settlement from the continent, but rather social changes that occurred before Bcdc wrote his history. Believing that distinct dialect areas must have resulted from distinct social groups, Bede may have assumed that 30П years earlier, the same social and tribal groupings had moved intact from the Continent. (37) PROBLEM 5.11: Here is a list of some of the Celtic words borrowed into English at various periods in the history of English, What do you conclude about the overall influence of Celtic culture on the Anglo-Saxon or on later society4-' (57) Pre-Anglo-Saxon: rfce (V i ugdom), ambikt (servant), dun (hill) Post-Anglo-Saxon: bruit (cloak), Ьапкис (piece of cake), gafeluc (small

WORDS: NiTIVh: AND BORROWED

55

spear), brocc (badger), carr (rock), luh (lake), ton (rocky peak), dry (magician), clucge (bell),aacor (anchorite), *tt№ (story),гЯе/(Шя hoard), cine (sheet of folded parchment). From I a tcr periods Irish: shamrock, leprechaun, galore, banshee, shillelagh, blarney, colleen, keen (wail). A few more than 40 borrowed into English. Scottish: clan, bog, plaid, slogan, cairn, whiskey. A few more than 30 borrowed into English. Welsh:crag,penguinisuU. A few more than 10 borrowed into English.

THE RISK OF ANGLO-SAXON ENGLAND

Whatever the sources of the dialect differences in these areas, political divisions developed to create the Anglo-Saxon Heptarchy, seven political spheres that often overlapped geographically: Northumbria, Mercia, East Anglia, Essex, Sussex, Wessex, and Kent. (See Figure 3.4.) The major dialect areas distinguished Northumbrian and Kentish, roughly coinciding with their

f) Г

Л ^

&•

f Clifiiti \

4

^WALES'

Yft.y> • \

О /

\

MERCSA

у

*—_

j EAST ) /ANGIA/

) J^v

,——1 "^ \ К llTJT "» Canterbury / X WESSEX ,•Vsussi^/"" CORNWALL p-^Ni^S™'^ 4

lLr

3.4. The Anglo-Saxon Heptarchy

S6

"WORDS AND MEANINGS

political boundaries. But Mercian included both East Anglia and Essex, while West Saxon covered both Wessex and Sussex. During the reign of Aethelberht (c, 552-616), Kent's political and cultural superiority eclipsed the rest of England. It* close commercial ties with the Frisian traders on the Continent are reflected in the large number of English coins found among [he remains of Frisian settlements and the number of Fnmfcish coins found in Kent. In the seventh century, Kenfs cultural dominance extended northward to the Humber, perhaps reflecting its prestigious position as a link to the commercial and intellectual life on the Continent, a life which at the time was probably still superior to England's. In the seventh and eighth centuries, cultural and political supremacy passed from Kent to Northumbria, then to Mercia, and finally to Wessex, where under Alfred (84^-899), there flourished a culture that surpassed anything on the European continent since the brightest days of Rome. In fact, were it not for the West Saxon rise to power and its accompanying literary flowering, we would have relatively few texts from before the Norman invasion. Except for some laws and charters, a Jittle verse, and a few translations of the Bible, the other O£ kingdoms have left us no great number of documents.

The Christian Conversion and a National The conversion of England to Christianity began during the reign of AethelberhC In 597, St. Augustine (?-604) became the first in an army of missionaries that wouid Christianize Kent in just seven years. And after contesting with Celtic Christianity (the Celts having already been converted by the Romans), they would win over Northumbria in 664. By 700 England could be called a moderately Christian nation. But Christianity did more to England than institute a new religion that encouraged new values. It re-introduced Latin and created monastic environments in which learning and scholarship flourished so richly that Europe was soon sending its students to Canterbury, Jarrow, and York. From the eighth to the eleventh century, southern England was one of the most advanced intellectual communities in the western world. During this period, England also began to develop a national character. From the eighth through the tenth centuries, Old English poetry flowered. Tl was during this time that the great poems of mixed Christian and Nordic themes were composed: Beowulf. The Wanderer, The Seafarer, the Caedmonian and Cynewultian poems. The efficient administration of a large institution, the church, provided a model for the secular kings in ihcirattempts to complement Rome's spiritual dominion of England with their worldly one. The organization of townships was roughly coterminous with parishes, each ruled spiritually by a priest who at first was often the chaplain to the local

WORDS! NATFVF. AVD BORROWED

57

thcgn (chieftain). The marriage оГ religious laws and secular enforcement created ;i governing institution of potentially great power and wealth. (2, 20ft, 223) ЗЛ2: Irt Problem 3.10 were listed some words borrowed from Latin during the Germ an ic continental period, before the invasion of Britain. Hud we listed [hem all and divided them into semantic categories, the p r o portions would have been very roughly as follows: Plants and animals: 307 0 Food, vessels, household items: 207Q Buildings, settlements: 12% Dress: 12% Military, legal: 9% Trade, commerce: 97 0 Religious, scholarly: Ъ% Miscellaneous: 5% Altogether, about 170 words were borrowed during this pre-OE period. Below are listed a weighted sampling of OE words adopted from Latin during the nest two periods: 404^650, and 650 to the end of the OE period, 400-650 prafost (provost), waster (town), cugfe (cowl), mentel (cloak), cist (chest), pxgel (pail), pott (pot), mwwc (monk), traht (text), cat! (cat), truht (trout), pert* (pear), gfcdene (gladiola), leahtrk (lettuce), xbs{ fir tree), senep (mustard), laser (.Larc, a kind of weed). 11 ii ii'

nulite (soldiers), cettiur {centurion), yndse (ounce), fenester (window), cluster (cloister), purs (purse), cxppe (cope), coc (cook), setae! (dish), rabbinn (to rage), scrofei (scrofula), credo (creed), discipu! (disciple), mmsse (church mass), papa (pope), xlmesse (alms), eretic (heretic), tnartir (шипут), organ (song), son (musical sound),sfol(school),pbiiosoph (philosopher), comet a (comet), bises (leap year day), ШНовесе (library), paitn (pahn), balsam (balsam), carte (dried fig), iilie (lilyX peoate (peony), ysope (hyssop), cancer (cancer), loppestre (locust), tiger (tiger), fenix (phoenix), camel (caniel). can you conclude about the nature of Lalin-OE contacts ? (7, 194) PROBLEM 3,13: Just as we can discover something about a culture from the words it ] m lost, so can we tell something from the words it often uses. Here 's a sampling of words that occurred very frequently in OE compounds, w ords made up out of two parts to express an idea no single word can, words in Old English like boccrseft (literally book-craft, or literature).

SS

WORDS ANfJ MEAiSTINUS

folclagu (folk-law. or law of the people), widsst (wide-sea, or ocean). What might you very tentatively speculate about a culture that used words like these

fairly often? 1.

sumof (summer), winter (winter), Sid (time), corn (corn), lyft (air), raht (night), №j*(ui(wood), wyrm (dragon), bhd (blood). 2. gold (gold), isen (iron), burg (dwelling), ham (home), sele (hall), medu (mead), bring (ring), win (wine), tar (learning), lead (song, poem), word (word), botf (book), gieip (boast, fame, pride), ftwrf (treasure), ceap (price, sale). 3. wtf (woman), wer (man), ^eow (servant), fread (people, nation), brodor (brother), ceorl (peasant), cyning (king), hfaford (lord), pegn (retainer). 4. hell (hell), heofan (heaven), cirice (church), crist (Christ), deofoi(dwW). 5. woh (error, iniquity), teona (injury), t&l (blame), syn (sin), sar (sorrow, pain), « 0 (strife, spite), morSor (murder), man (evil deed), bealu (harm), cweahn (death), dead (death), Awfe (error, heresy). 6. eufie (ea iy), wynn (joy), gliw (pleasure, sport), 7. zsc (spear), йе<а1н (battle), borrf (shield), camp (battle), here (amy), Jn-'/r? (army, people), wspen (weapon), guS (war), 8. sa: (sea), scip (ship), и?г/ег (water), ;jff (wave). 9. /^/(praise), аде (victo ry ) t i™/KOBi.FM ЗЛ4: What might you guess about a culture that has a kinship system with the following terms? Compare it with Modern English. (Review Problem 2.13,) jseder (father), Mudor (mother), sunn (son), detoor (daughter), broSor (brother), sweostor (sister), fsdera (paternal uncle), earn (maternal uncle),/аЭе (paternal aunt), modrige (maternal aunt), tukterga (brother's son), swigra (sister's son), fsikrencnosl (father's kin), jnodorcym (maternal descent), wmpnedtutrtd (male line), mfttund (female line).

The Danish Invasions But as the West Saxons in particular were creating their cultural golden age, their bin from Denmark, more comfortable in their dragon ships than the now-landed Saxons, began in the late eighth century to find easy targets among the rich monasteries along the eastern and southern English coast. Finding how defenseless Northumbria and Easi Anglia were, they escalated their raids to a full scale invasion from K5O to 878. By Lhe time the Danes had overrun, occupied, and begun to colonize large portions of eastern England, only Wesbcx was able to resist them successfully.

WORDS: NATIVE AND BORROWED

59

Canterbury

Figure 3.5. The Uanelaw In Я78, King Alfred finally defeated the Danes in the Battle of Ethandun. Jn the subsequent Treaty of Wedmore, the Danes agreed to remain east of a line running roughly from north of Chester to London, an area called Danelaw (see Figure 3.5), in which the Danes were free to live as Danes under Danish law. (A second Danelaw, which, if anything, would prove to be even more significant to the future of England, was established in the Prankish Kingdom on the continent directly across from England. It was called, after the Northmen or Normans, Normandy.) Having united at least half of England, Alfred, as noted above, set about creating almost single handedly a prose tradition by his various translations from Latin into Old English, both by his own hand and at his direction by others, by his initiation of the Anglo-Saxon Chronicles, and by the importation of European scholars and books. He also founded the first public schools, creating a literate audience for the literature as well as an educated class able to administer the growing bureaucracy of a growing state, Iftfact, AI fred organized A n glo-Saxo n go vcrn m ent so we! I th at the Danes in their Drmelaw, who were less well organized than the English, were finally unable to withstand the unified political and military attacks by Alfred's son, Edward ine Elder (c. 870-924), and his son, Atheistan (c. 895-940). Tn 957, Edward's grandion, Edgar (У44-975) became ruler over Northuinbria and Mercia. And in 959, upon his brother Eadwig's death, he became king of

60

WORDS AND MEANINGS

Wessex as well, ruling an England that extended from the Tyne south and west to the Welsh mountains, roughly the England we know today. Here are two samples of OE literature with literal translations. The first excerpt is from The Venerable Bcde's story of Caedmon, a cowherd at a Yorkshire monastery who in a dream was given the divine gift of creating religious songs. Below are the opening lines of that story. The second is from the "Battle of Brimanburg," a poem included in the AngloSaxon Chronicle from the year 937, It narrates a victory of the Anglo-Saxons over the Scots and Danes. (The text has been normalized somewhat to make spelling more regular,) In ocosse abbudissan mynstre wis sum brooor syndrigiice mid In this abbess's monastery was a (certain) brother especially with godcundre gife gemsered ond geweorSad, lor pon be gewunade heavenly gift famed and blessed, for he accustomed (was) gerisenlice leoo wyrcan, pa $e torcfisestnisseond to arfsestnisse appropriate songs (to) make, which to piety and to virtue belumpen; swa Ssettc swa hwtet swa he of &odcgndum stafum }>urfa pertained; so that whatever) be of heavenly letters through boceras geleornodc, J?set he sfter medmiclum Гэке in scopgereorde scholars learned that he after moderate time in poet's language mid ha msestan swelnisse ond inbryrdnisse ge&iengde, ond in with (he most sweetness and humility adorned, and in Hngliscgereordc wei gewoi'ht Гоф brohtc. Ond for his Icu^sotigum English-language well worked forth brought. And for his poemsongs monigra mtmna mod oft to worulde forhogdnisse ond to many (of) men's minds often to world contempt and to ge^eodnissc |>ffis heofonlican lifes onbsrnde wiron. association (of) the heavenly life inspired were. Her ^pelstan cyning, / eorla drihten heovna beahgifa, / and his broSor eac, Eadmund «Seting, /ealdorlangnctir geslogonffitsa^tce / sweorda ecgum yinbe Brunanburh; / bordwcall clufon, heowon hcaQolinde / hamora lafum, eaforan Eadweardes; / swa him

Here, Athelstan king.earis' lord warriors' rlnggiver and his brother also, Edmund nobleman, Lifelong glory won at battle (with) swords' edges around Brunanburg; shieldwalldove, hewed warlinden (with) hammers' leavings sons Edward's; so (
geaeftele w s s

Tram cneomagtim. / pxt hiffitcampe oft wi6 lafira gehwone / lnnd cal&odon, riord and ha mas. / Hellene! crungon. Seotta kodc / and scipflotan, f«gc feoLIon; / feld dennode secga swate, / *i]s]san sunne upp

from antes!ors that they at battle often against foes any land protected treasure and home. Army died, Seotfisb people and sailors, doomed fell; field streamed (with) warriors' sweat since sun up

WORDS: NATIVE AND BORROWED

on morgenlid. / rtiffire lungol, glad ofef grundas. / G odes catidel beorht, eccs Drihtnes, / o3 sio stuck gestcafl sah to setlc. / ржг Ijeg sccg monig garum Idijfruiidan, / guma NorSerna ofcr scyId .scoten, / s w уIce Scyttisc eac werig wigesssd.

61

on morningtime, famous star, gli ded over groun ds, G odhs candle bright, etem&J Lord, until Ihe noble creation sank to scat. There lay warrior many {by) sowars killed, men Northerners' оver shiei d shot, so Scot also weary (of) battle sa tco\

Here, King Allielstan, lord of earls, ring-giver of warriors, and also his brother, Edmund the nobleman, won lifelong glory in battle with their swords around Hiunanburg. The sons of Edward1 clove shicldwaHs, hewed linden-shields with the leavings of the hammer (swords). It was natural from their nohle ancestry that they in battle often pro led ed any land, treasure, and home against any foes. The army fell, Scottish people and sailors fell doomed. The iield streamed with the blood о Г warriors from the tiime tlie sun rose up in the morning, that famous star, the bright candle oi' God, the Eternal Lord, glided over the land, until the noble creation set. There lay many a warrior, killed with spears, Norseman shot over their shields, Scots too, weary and sated with battle, PROBLEM 3.15: You have already seen two foreign influences on English, Latin and Celtic, Here are some Danish words that were borrowed into English. (Some of them have been subsequently lost.) (I) What kind of contact did the Danes and Anglo-Saxons have? Before A,D. 1000 barda (beaked ship), cnearr (small warship), 1ф (fleet), ha (oarlock), arrest (battle), ran (rapine), mctl (action at law), hold (freeholder), wapeniake (an administrative district), husting (assembij1)AfterlOOO Nouns: band, booth, bull, dirt, down (feathers), egg, fellow, freckle, kneel, kid,fe£>link, reindeer, reef{of sail), scab, scales, scrap, sear, sister, skin, skirt, jfcy, snare, steak, swain, window; birch., boon, gait, gap, guess, loan, race, rift, score, skill, slaughter, stack, Shrift, tidings, trust, waul, gift. Verbs: call, crawl, die, get, give, lift, raise, rid, scare, take, cast, clip, crave, droop, fiii, gape, kindle, nag, scowl, snub, sprint, thrust. Adjectivesxflat, loose, low, odd, tight, weak, awkward, ill, meek, seemly, sly, rouen, tattered, muggy. Pronouns and iither words: they, their, ihetn, both, same, though, till, (2) In the Oxford Universal Dictionary, there are approximately 30 or so ds beginning with sk-lsc- of Danish origin still in our active vocabulary. in the English Dialect Dictionary, there are over a thousand simple words

62

WORDS AND MEANINGS

beginning with sc-Ltk-. What do you conclude about the durability of borrowings in non-standard dialects? What dialect area wouid you guess they are from? 0) Is the part of speech significant in the borrowings listed above? Why?(] I) (4) Review Lhe borrowings from Latin and Celtic. What differenceis there in the tone of words like iiraop, scare, nag, muggy, and/recite on the one hand and, on the other, words in Problems 3.10 and 3. ] 2? When the Danes were Forcibly brought into a not very solidly united England, it did more than begin the political unification of the land. While surpassed by few in their military zeal, the Danes were equally skilled in commercial affairs and in boning their legal points to a fine edge in their Thing, or meeting of ciders. Earlier, in Anglo-Saxon England, a violation of the law was often followed by a blood-feud. The strong sense of Danish legality combined with the increasingly strong English local government made crime less a private question of one individual compensating the kin of an injured party with a hnc eailcd werglid than something to be dealt with by those who spoke in the name of the local thegn, the king, and God. Real, immediate, practical political power, though, was still exercised by the local thegn and priest. The thegn supplied land in return for his people's labor and its fruits, protected them, and dispensed justice. In this system lay the seeds which would grow into a social structure powerful enough to shape the English social system for hundreds of years after the Norman invasion. The ploughman ploughed and the thegn governed and fought. And if this meant a more stable and productive society for everyone, it also rneaTit less freedom for Lhose who pushed their ploughs. Because the power of the throne was not yet strong enough to reach down to the individual thegn and churl, it could not enforce a single standard of justice. Long after the putative unification of the kingdom, the Danelaw continued to exercise a considerable degree of independence in its own affairs. As a consequence, England was still divided in spirit when new Danish raids broke out, particularly during the reign of Ethel red the Rede less {c. 968-1016), a rather incompetent ruler. Since the union beLween the peoples was not an easy one, the Danes did little to resist the raids against Wesscx, an area outside their Danelaw. After Elbelred's death, Saxon and Danish England engaged in a brief civil war before Cnute (c. 994-1035), a Dane, defeated Wessex in 1015. He and Edmund Ironside (c, 980-101 d), Ethelred's son, briefly divided the island between them, and when Edmund died in ]0!6, Cnute became the king of the entire kingdom. As it turned out, not only did Cnute rule wisely, but his accession to power opened southern England, particularly London, to Danish businessmen and traders, making London an even more powerful and cosmopolitan city than it already was. (223) PROBLEM 3.16: Words borrowed from the Danes do not begin to occur

WORDS: NATIVE AND IWIKKUWED

63

frequently in English texts until the Middle English period. What might be one explanation for this 7

The Rise of London

Because London figures so centrally in the development of a standard English, we have to account for its unique position in English history. Although some sort of settlement undoubtedly existed on the Thames before the Romans arrived, it was during Roman times that London began to develop into the iirst city of the country. The spot on which London iits was the onlypiece of hard ground on the northern side of the Thames that afforded a solid bridgehead for roads coming from the Kentish towns and a suitable landing place for ships coming up from the English Channel. Because half the roads built also converged on London Bridge, London was long destined to be the future со m in ere i ill center of England, Under the Saxons of Mercia, London declined somewhat in prestige since commerce with the continent was not as great as during the Roman occupation or Kent's ascendancy. But though of little relative importance, London maintained a measure of independence from both Mercia and Kent, and when Alfred settled with the Danes in 87S, he managed to exclude London from the Danelaw, [hereby preserving its Saxon character. Because it was the main entry to the heartland of England, he fortified it and encouraged its growth in order to defend it from the Danes, Then when Cnute assumed the throne in 1016, Danish merchants became some of the leading tradesmen and citizens. Before the end of the eleventh century, London had become the most important commercial city in England, populated by a variegated and sophisticated people, many from continental Europe. Always more powerful than her official status would suggest, London finally regained the status it had held in Roman times. By 1066, her population stood at perhaps 14,000— several thousand more than the next largest city, Winchester, and perhaps 6000 more than the estimated 8000 population of York. After the Norman Conquest in 1066, William the Conqueror was crowned near London in Westminster. But when he built a residence, he moved from inside the walls of the City closer to his new Cathedral at Westminster, two miles away. In this single action, he made the few miles encompassing Westminster and London the political, commercial, and cultural center of the land. At the same time, by living outside the walls of London, he helped preserve its independent political, cultural, and economic spirit. London's later political and cultural strength eventually resulted in its dialect becoming the prestige dialect of the land. While Alfred and his descendants made tlieir court in the South, West Saxon was the dialect of

64

WOKL1S AND MEAN1NUS

English in which the major literature and law was written. When the Norman Conquest ended Wessex's ascendancy and reduced the flow of literature written in English to a trickle, the only prestige dialect among the upper class was Norman French. When English hegan to re-assert itself three centuries later, it would be the dialect of London, of the East Midlands, that would eventually become the national standard. This raises a difficulty in studying the history of English because West Saxon, the earlier prestige dialect, and Mercian, the OE dialect that would father East Midland, were different in some important ways. When we study OF., we study West Saxon, hecause that dialect was used in the great preponderance of OE texts. But standard MudE stems not, ultimately, from West Saxon but from East Midland speech.

Chapter 4 FROM MIDDLE ENGLISH TO MODERN ENGLISH

THE NORMAN CONQUEST

After the migration of Ihe Anglo-Saxons in the fifth century from the Continent to Britain, the Norman Conquest was the single most significant event in the external history of the English language. The Danes added many words to the vocabulary and a very few grammatical changes, but because Danish was somewhat like OE to begin with, its impact со aid not have been as influential as Norman French. The Norman Conquest has led to a language that is qualitatively dirt ere nt from what it was before JOfifi. The groundwork for Norman dominion began long before 1066. During Cnute's rule, Edward the Confessor (c. 1002-66), son of tithe!red and Emma, daughter of a former Duke of Normandy, returned from a long exile in Normandy, strongly influenced by the Norman clergy and more interested in being л monk than a king. In facL, when he assumed the throne in 1042 he probably spoke better French than English, Because of his background, he appointed Normans to high positions in the English ciergy and government, and by allowing Normans access to London, he introduced them into Lnglish commercial life. Moreover, he made no attempt to unify the isiand, strengthen 65

66

WORDS AND

its defenses, or redress the growing imbalance of power exercised by a few earls jcakms uf one another's power. When Edward died in January of 1066, a seriously divided and weakened England fated the additional problem* brought about by his fidelity to his monkish vow of chastity. He left no direct heir. There were, however, two possible oblique heirs. One was Edgar Atheling (c. 1050-c. 1125), considered by most to be at 16, too young to assume the throne. The other was Harold IL 0022-66), a more distant relation but one with the blood of King Olaf of Sweden. He was finally chosen by the Witan (the king's council) to be king. Then only eight months later, lingland found herself onee again attacked by Scandinavians: by Harold Haardraada, King of Norway, and by William (1027-&7), Duke of Normandy, who believed he had the only legitimate claim to the throne. In 1051, Edward had promised him the crown and in 1064, William had made Harold II promise that he would help him, William, gain the throne. On September 25, 1066, at Stamford Bridge, Harold IT defeated a force led by Harold Haardraada of Norway. But four days later William landed at Pevensey. On October 14 at Senlac, about nine miles from Hastings, the English and Normans fought until nightfall. The Normans drew the English from their impregnable position with a pnetended flight and defeated them. An arrow through the eye felled Harold. Thus England became a part of the Latin-influenced rather than Scandinavian-influenced life of Europe. In a few years, through persuasion and massacre, William largely erased the old political distinctions between the Danelaw and the South. England was, though brutally oppressed, at last politically united under one sovereign king, William simultaneously deprived the English earls of their power by breaking up the last vestiges of the traditional mini-kingdoms of Mercia, Wessex, and the others, and replacing them with a system of shires and baronies with his own local officials, all Norman. As might be expected, William also replaced the English nobility with Norman barons and the native church establishment with his own Norman clergy. By 1076, not a single English earl remained. Only three of 21 abbots were English. Under William, incipient feudalism became a harsh social reality. The half-free Danish freemen in their Danelaw lost much of their freedom to choose which lord to follow or in some cases, even to have no lord at all. For now both freemen and serfs had to serve some lord.

THE LINGUISTIC CONSEQUENCES In retrospect, the linguistic consequences of this invasion and colonization arc easy enojgh to recognize. Although certain grammatical and phonological changes may be attributed to the Norman Conquest, it is in vocabulary

FROM MIDDLE ENGLISH TO MODEftN FNtiUSII

67

and semantics that the changes have been the greatest. The enormous influx in Ml: of not only French but also LaLin words certainly would never have occurred without the Conquest But if the effects are easy to discern, the precise social milieu in which they occurred, die specific reasons for their occurrence, are less easy to reconstruct. pnoBLENf 4.1: This problem will require cooperative effort. (1) ScTect three or four different kinds of twentieth-century prose texts: scholarly, popular, technical, comic books, etc. lie certain to get a contrast. In a 250-word sample of each kind, tabulate the percentage of native to French/Latin to alt other words. Simplify the task by courting as native words all prepositions, articles, pronouns, auxiliaries (may, must, be, time, etc.) without looking each one up. Would the percentages change if just nouns, verbs (excluding/w), adjectives and adverbs were counted? How would your conclusions differ? (21 Select three dictionaries: a very short, paperbound лпе; a larger standard desk-we one; and a larger reference dictionary such as Webster's Third or The К and urn House Dictionary of the English Language. Cojnputc the percentages of native English words to French-Latin and all others under 50 randomly chosen words for each of six or seven letters. What is wrong with this method 7 What variables make the problem more complex than meets the eye? (3) What are the significant differences among the kinds of texts in question 17 What are the significant differences among the kinds of dictionaries? What are the significant differences between these two ways of counting percentages of word sources? How do you account for these differences? What would have happened if you had counted the words under V? The words under A? under К 7 PROBLEM 4,2: If we group the vocabulary into the first most frequent thousand words, second most frequent thousand, third most frequent thousand, and so on, then compute the percentage of native versus borrowed words in each of these groups of a thousand, we find figures such as these: Decile

1 2 3 4 5

б 7 8

10

Kflfilisli

83% 34 29 27 27 27 23 26 25 25

French

Latin

Danish

Other

И7о

2% 11

2уо

1% 7 10 10 8 10 13 13 15 14

46 46 45 47 42 45 41 41 42

14 17 17 19 11 18 17

IS

2

1 1 1 2

г2

2 1

68

WHUDS A N » MliAMTNGS

(The "other" group includes mostly mixed or doubtful words, or words that only might be assigned to English, French, or Latin words. Only Dutch among "others" exceed* I percent in any of the deciles). When ;ill the words in running text arc put into one group, the percentages are as follows: English: 78.1; French: 15.2; Latin: 3.1 J Danish: 2.4; other (Greek, Dutch, Italian, Spanish, German, tie): 1,3. Comment, These data were compiled from several thousand business letters. (18I) Comment. Here, in considerable detail, are sets of demographic statistics, social anecdotes, facts about official documents, and so on that may be relevant to understanding how French and English were used at various levels of English society. Some historians have claimed that French virtually ousted English at all levels, oral least that almost all Englishmen were bi-Iingual. Others have claimed thai only a thin layer of the nobility habitually used French and that the vast majority of English were effectively monolingual. As a background, here are some sketchy over-generaiinations about Norman society: Ruling all were the king and his court. Below him were his barons, and below them were "enfeofled" knights who were granted the right to land by the king through his barons in return for service to them. Under the knights were a middle-management layer of officials: reeves, bailiffs, stewards, and so on who ran the estates and local affairs. Below them were the great mass of Hnglishmen and some Norman peasants: the churls, the villeins, the laborers. In the towns or boroughs, the great mass of citizens were skilled and unskilled laborers. Above them were the tradesmen and businessmen. A third less statistically significant element of the population was the clergy. Some axioms: 1.

2, X 4,

When social order is fluid, upwardly mobile lower classes are more likely to behave in ways they believe upper middle classes behave than when the society is rigid. Social order becomes more fluid during periods of social upheaval. In areas of high population density, contact between social classes is greater than when they are thinly distributed. Contemporary observers of a social scene are more likely to comment on the unusual than the usual.

The initial numbers in the data below refer to century and decade. Following each set are Problems based on the data.

Eiconomic-demogrdphk LstimaUs

11.7

Jn 1066, the Lnglish population stands between 1.1 and 1.7 million.

FROM МШСН-Е ENGLISH TO MODERN ENGLISH

<#

the great majority south of a line running from Bristol to The Wash, the deep inlet of the North Sea between Norfolk and Lincoln. The Norman invasion force is estimated at between 5000 and 12.000 men. After the Conquest, when land is distributed to Nor тип knights, there are about 5000 Norman knights enfeofled (granted land) out of 10,000 smaller landlords. The other 5000 are presumably English. 11.9 On the basis of the Domesday survey in 1086, a kind of census and economic survey ordered by William for tax purposes, the English and Norman population is still estimated at between 1.1 and 1,7 million by modern scholars. They are distributed roughly as follows: Rural, at least 85 percent: vaguely or certainly urban, no more than 15 percent, probably less. Among these is the clergy at about .5 percent. The population of London is about 14,000. The other large towns constituting a large part of the urban population are Winchester, Lincoln, Norwich, Gloucester. York, and Canterbury. The Fast Midlands, the area north of London, is the most densely settled part of England and the- least ravaged by the Invasion and its aftermath. The north is ihc least populated. The heaviest concentration of Norman population is probably in the commercial towns along the southern and eastern coast, though a ^real many settle inland. 13,1 In the twelfth and thirteenth centuries, the East Midlands grows rich exporting corn and wool. 13.6 The richest cities in England are London, Lincoln, Winchester, Canterbury, Worcester. Oxford. York, Northampton, and Norwich. J3.I0 Before the end of the thirteenth ceniury, the curlier immigration from the south and southeast shires ends. Before 1300, 38 aldermen and IK sheriffs in London can be identified as coming from southern counties. The Fasi Midlands contributes 10 aldermen and 7 sheriffs, 14,1 I n 1300, the Lond о п pop ul a ti on is st i 11 abou U 4,000, 14.5 In 1348, just before the Black Death, the total population of Lngland is between 3.1 and 3.7 million. Before the century is out, the Black Death will eventually kill 30 to 40 percent of the English population, 14.5 In 1349,47 monks who die of the Black Death in one monastery are replaced by uneducated Englishmen. Large numbers of uneducated men whose wives have died of the plague turn to holy orders. 14.7 Tn the first half of the fourteenth century, emigration from the East Midlands and the North into London increases. Between 1300 and 1365, southern immigrants contribute 32 aldermen and 7 sheriffs. The East Midlands contributes 33 aldermen and 10 sheriffs. Northern areas contribute 5 aldermen and 2 sheriffs before 1300. 10 aldermen and 2 sheriffs between ] 300 and 1365.

70

WOUDS AND MEANINGS

14.S 15.1

By 1375, after more plague, the English population stands at about 2,250,000. By 1400, the population of England falls to about 2.1 million. The heaviest mortality lias been among the very young and the old, particularly in the crowded monasteries.

PROBLEM 4.3: If we bad only these data and n o ! in guistic evidence whatsoever, what predictions might we make about the prestige of Norman French, its general influence across the entire island; whaL area of the country would become soci ally most prestigi о us, incl ud in g i Ls dialect ? Political Events 11.7 11.8

In ] G6G, Willi am invades and conquers Engian d. In 1075, 13 of 21 bishops signing the decrees of the Council of London arc linglish. By 1087 only three are. 11.9 In 1086, there are 190 barons, all Nurmans, No English earls survive the suppression of the English nobility. 11,9 In 1087, William I dies; William 11 becomes ting, 111 In 1100, William II dies; Henry I becomes King. 12.4 In J135, Stephen succeeds Henry I. 12.6 In 1154, Henry 1) succeeds Stephen. 12.9 I n 1189, Richard 1 s ucceeds Henry II. 12.10 In 1193, London becomes the first town to have its own mayor, 12.10 l a 1199, John succeeds Richard I. 13.1 In 1204-05, Philip of France, whose court is in Paris, seizes the Norman estates of several powerful Anglo-Norman barons plus all the French lands of knights who live in England, Anglo-Normans must now choose whether lo become Englishmen or Frenchmen. King John is left with possessions only in the south of France 13.2 Jn 1215, King John gives Londoners the right to elect their mayor. 13.2 In 1215, King John assents to the Magna Carta. 13.2 In 1216, Henry 111 succeed s John. 13,4 In 1233, because of his close connections with the south of France, Henry 111 dismisses the native officers of his courl and replaces them with French from Poitou, Two thousand French knights and soldiers from southern France are placed in charge of castles and lands, oppressing natural English-Norm а г subjects and noblemen, 13.4 In 1236, when Henry III marries Eleanor of Provence, a new influx of southern French arrive in England to be given more power and lands. 13.5 I n c . 1246, a third influx of southern French occurs when Henry 11 I s mother dies. Having married a southern Frenchman on the death of

FROM MIDDLE ENGLISH TO MODERN ENGLISH

71

John and borne him several sons, she leaves Henry with half brothers and their daughters to provide with lands. 13.6 The Barons Wars from 1258-65 sets native English against French usurpers in England; the latter are finally driven from the island. 13.8 In ] 272, Edward ] succeeds Henry ITI. 14.1 In 1307, lid ward II succeeds Edward I. 14.3 In 1337, Edward 1ТГ succeeds Edward II. 14.4 Jn 133 7, th e H und red Yea гкп War wi th Fra n cc hegi ns. 14.5 In 1346, the English gain a great patriotic victory over French, at Crecy. 14.6 In 1356, English patriotism is aroused again with a victory at Poitiers. 14.8 Tn 1377, Richard 11 succeeds Edward III. 14.9 In 1381, many peasants refuse to remain tied to their feudal Lords any longer when they can sell their labor at higher prices. The Peasants' Revolt occurs. 14J0 I n 1399, H en ry IV su cceeds Richa rd 11. 15.2 In 1413, Henry V succeeds Henry IV. 15.2 I n 1415, E n gl ish gai n an other great pa triotic victory at Agincourt. 15.3 In 1422, Henry VI succeeds Henry V. PROfiLEM 4.4: (I) Which political events would probably have had an impact on the пне of French in England? (2) Are the- consequences of any of the political events reinforced by demographic events, or vice versa ?

Written Kriftlish, French, and Latin in Official Texts and Legal Documents

(Note: All "firsts" should be understood to mean " earliest known,") 11.8

In 1070, William issues writs in English, some in Latin, none in Norman French. 11.9 After 1030, writs in English virtually disappear. William has laws written in Latin. 12 Throughout the twelfth century, Old English laws are ordinarily translated into Latin. 12,1 Until 1109, Lhe Annaies AnRlo-Saxonici Breves, compiled at Canterbury, are written in English, thereafter in Latin. 12.5 An Anglo-Norman charter appears, the first known in the language of the Normans. 12.6 First laws written originally in Norman French dated 1150 от a bit earlier. 12 rt Jn 1154, the Peterborough Anglo-Saxon Chronicle, the only native language chronicle that had been maintained, lapses. 12.9 Only a few charters and other documents appear in English until the reign of Richard 1(1189). Latin is by far the more common language.

г

72

WOKLKS ANfJ MhAMNCJS

13.2

The earliest extant deed written In Norman French dates from 1215, 13.4 Л Rtiyal Charter in 1233 says that English law terms have been translated into French so that "everyone will understand,'" 13.6 In 1258, the first Letters Patent, official documents, appear in French, Previously, they were in Latin. 13.6 In 1258, a group of barons and upper-class Englishmen win from Henry III the Provisions of Oxford, guaranteeing certain rights to the barons. It is set forth in French and English—the first such public document in English in almost 200 years, 13,8 Before Edward I, most petitions and bills to Parliament are in Latin, after that time in French. First Parliamentary writ in French in 1274-75; first petition to Parliament in French in 1278. 13.8 From 1275, laws are normally drawn up first in French, 13.9 By the close of the century, French is used in most official documents. But petitions to Parliament in French are translated into Latin so t h a t " all may understand," 13.10 In 1299, an order regarding the use of the forest is published in English. 14.1 In the closing years of Edward I's reign, French first appears in Privy Seal documents, official documents of less than major significance. 14.1 Throughout fourteenth century, it is common for deeds, bills of sale, contracts, and so on, to be written in French. 14.5 Before the middle of fourteenth century, most petitions and bills to Parliament are in French. The earliest petition (to the chancellor) in English appears in 1344. 14.5 Tn 1345, the Pepperers Guild of London uses English in their guild records for the first time, the first guild to do so, 14.6 First wi 11$ hegi n to appear in French. 14.8 Earliest deeds written in Middle English appear near the end of Edward Ill's reign. 14.8 In 1376, first private legal instrument written in Middle English appears, 14.9 In 1383, first will written in Middle English is filed. 14,9 In Ш 6 , the London Mercers Guild sends Parliament, first petition written in Middle English. 14,9 In 1388, in answer to Parliament's request for information, the English guilds respond mostly in Latin, but more responses are in English than in French. 14.9 Between 1369 and 1384, French deeds registered on Close Rolls, official records for private transactions, average about 29 a year. 15.1 During reign of Henry IV, English begins to supplant French in royal documents.

FROM MIDDLE ENGLISH TO MODERN ENGLISH

73

15. J

In 1404, Lnglish write to French government in Latin and ask them to reply in same. 15.2 The last official Letters Patent are written in French during reign of Henry V. 15.2 Between 1403 and 1418, French deeds on Close Rolls average four a year. 15.2 in 1413, Henry IV dies, leaving first will of a monarch written in English. 15.3 In 1422, first Privy Seal document in English appears. 15-3 Jn 1422, London Brewer's Cjuild, praising the English language, begins to keep records in English, 15.3 Until 1423. most petitions to Parliament are written in French; after this date, they are frequently in English. 15.4 Last wi IL i n French regis tercd in Cl ose Roll s appears in 1431. 15.4 Between 1419 and 1434, French deeds registered in Close Rolls average one a year: none after 1434. 15.4 About 1430, several towns and guilds begin translating their official documents from French into English. 15-4 In 1437, French is discontinued for use in Great Seal documents, documents of the greatest national importance. 15.5 Between 1433 and J443, Parliamentary petitions in French decline markedly, none occurring in French after 1444. 15.6 One of the iast legal instruments in French is dated aroitnd 1450. 15.6 By a bon 11450, m os t reco rds of tow ns a nd gu i ids are i n En gl ish. 15,6 By the middle of the fifteenth century, Parliament needs a Secretary in the French language. 15.9 In 1485, English occurs with French in Statutes of Parliament. 15.9 By 1489, French disappears entirely from Statutes of Parliament 15.10 By the end of fifteenth century, French к used only in Jaw cases— English in almost all other kinds of documents (except for Latin scholarship and ecclesiastical records). 18.4 In 1731, records of lawsuits are regularly written down in English. РШМ.ЕМ 4.5: (1) Which of these events reflect political events? (2) Plot the shift from OE to Latin to French to English in official documents. (3) Even Irom this very sketchy data, can you determine the crucial periods? Written English, French, aw) I a (in in Social Commuii [cation and Literature 12

Early twelfth-century translators say they translate from Latin into French ко that "thecommon man " will understand. с 12,5 A Latin vocalulaiy is produced with words glossed only in French. 12-13 Much literature is written in French for literate Englishmen in the twelfth and thirteenth centuries, relatively little in English,

74

WORDS AND MEANINGS

12Л

Beginning about 1175 and continuing through the middle of the thirteenth century, many passages in works written in French for F.nglishmen suggest they are written in French so that " a l l " can understand. 12.10 The Owl and the Nightingale, one of the best debate poems in ME, appears. i 3.1 Lay am on's Brut is written, one of the first romances in ME. 13.1 The Chancellor under Richard T is charged with having flattering songs written about himself to be sung in the streets of London by minstrels and singers from France. 13 Phrase books for Leaching French to Englishmen for business purposes begin to appear. 13.1 Three young upper-class women art advised to do their reading in either French or English. 13.2 The earliest extant letter in Anglo-French in Middle English period dates from J215. 13.3 Л Latin vocabulary is produced glossed in both French and English. 13.5 Beginniitga round 1245, m any s tateme nts a ppea r in N о rm an French works written in England decrying the decay of the Norman French language. И.6 In 1250, a short treatise in Latin appears, explaining the French verb. 13.7 After the middle of the thirteenth century, English literature begins to revive. Havelok the Dane, one of the earliest romances about an English subject, appears. 13.8 About the end of the thirteenth century, educated writers begin to apologize for their "false French," i.e., incorrect French. Mistakes in grammar and awkwardness in style begins to appear. 13.9 In the last quarter of the thirteenth century, most private letters to and from the nobiiity are in French, the rest in Latin. 13.10 AD admirer congratulates Giraldus Cambrensis on his writings, but regrets that Cambrensis did not write in French instead of Latin, for he then would have been more widely read. 13.10 At the eitd of the thirteenth century, the opening lines of The Romance of Richard the Lion-Hearted include: Lewede (unlearned) men cum (know) Ffrensch поп j Among an hondrydvnnefris (scarcely) яч14.1 About И00, these lines appear in a metrical homily: fart hi will of my pavert / Schau sum thing thai Ik hafin hert, / On Ingelis ions that aik may j Understand qual I wilsay. /. . . Bet al men can nohf, I-wist / Understand Latin and Frank is. 14 In the fourteenth century, the overwhelming majority of private letters from nobility, religious houses, and so on are in French, the геь1 in Latin.

FROM MIDDLE ENGLISH TO MODERN ENCrl.TSH

14-1

14.1

14.3

143

75

About 1303, in the preface to Handlyng Synnt, appear the lines: For fewde (unlearned) men у wutyrtokel On engtyssh tunge to make ftys hake. About 1300, in the Cursor Mundi, appear the lines: pis ilk bok es translate I Tnio Inglis JOftg to rede (counsel) / For the lore of Inglis fede (folk), / Ing/is lede of Ingland, / For she communal (to) understand. About 1325, the opening lines of Art hour and Merlin include these: Man: noble kh haue ysei$e {seen) I Pat no Freynsdie coupe {can) seye. About 1325, these lines occur in William of Nassyngton's Speculum Vitee: In English tonge I sehal 30W telle, jif 3e wyth me so longe wil dwcllc. No Latyn wil I speke no waste, But English pal men vse mast, pat can eclie inan vnderstandc, Pat is bom in Ingclande; For put langage is most chewyй (shown) Os wel among lered (learned) 05 kwyd (unlearned). And somiTie can (know) Frensche and no Lalyn, fat vseel han towrt (court) and dwcllcn herein. And sonime can of Latyn a party, pat can of Frensche but Jebly; And sommc vndcrstonde wcl Englysch, Pat can noj^er (naithei1) Latyn nor Frankys, Ло^е [ered and к wed, olde and jonge, AUc vndcrstonden english tonge.

14.5

14,7

By the middle of the century, -virtually all correspondence is in French. In the last half of the century, writers make more and more grammatical mistakes. Between 1360 arid 1400 appear the most important Middle English work* of literature: TheFeerl, a dream vision; Sir (i a wain and tli-c Green Knight, a romance; and The Vision #f William Concerning Piers the Plowman, a combination of satire, social commentary, religious allegory, and homiletic. All of these were part of the alliterative survival in the. late fourteenth century that drew on. Old English verse forms. From about 1360, Geoffrey Chaucer (c, 1340-1400) begins creating the greatest English literature of the period and some of the best of all time, including The Canterbury Tales. About tins time, the Alliterative Morte Arthurs appears, a romance about King Arthur, written in the native alliterating verse that goes back to the OE tradition.

76

WOBBS AMD MEANINGS

14.8

About 1370, the auth or of Piers Plowman wri tes: Gramer, the groundc of al, bigyleth now children; / For ib none of this newccierkes, who so nymelh hede, / That can versifye faircneformalieh enditen; / Ne nou3t on amonge an hundrcth (hat an auctour can construe / Ne rede a Leltre in any Langage but in Latyn or in Englissh. (B-test, XV, 365-369)

14.9

14.9 14.9

About 1375 to 1425 or so, English writers switch back and forth between Hnglish and French in their correspondence, a phenomenon uncommon either before or after. John Wyciiffe's (c. 1320-84) translation of the English Bible is published posthumously. In 1385, Thomas Usk writes: L*:t then clerkys endytcn in Latin for they have the propertee of science, and the knowings in that fecultee: and let Frenchmen in their Fnenche also endyten their queynt terms, for it is kyndcly to (heii1 moutbes: and iet us she we our fantasy cs in such words as we lerneden о Г our dames tonge,

14.10 The first known private letter in Fnglish dates from 1392-93; the next from 139914.10 Until the end of Richard tYt reign (1399), manuals for letter writers with model letters in French abound. They do not instruct in the basic grammar of French but in socially correct writing. 15,1 In 1400, a writer to Henry fV apologizes for writing in English rather than French or Latin, but, he says, he understands English better, 15.1 About 1403, Richard Kingston, Dean of Windsor, writing to Henry IV: Jeo prie a la Benoit Trinite que vous ottroie bone vie ovc (resentier sauntee a trcslonge durre, and sends 30we sonc to ows in helj? and prospcrilee; for, in god fey, I hope ю Al Mighty God that, 3ef ;e come jouve ownc persons, зе schullc have «he victoric of alie joureenemycs.... 15.1 15.1

J5.3 15,4

Many upper-class writers write some letters in English, others in French, seemingly indifferently. I n the earl уfifteenth ce ntu ry, Jo hn Ba rton i n Doiiet Francois adv ises that adulLs should learn French so they can (1) get along in France> (2) understand laws, (3) write polite letters to other upper-class Englishmen. In 1424, the earliest letters in English from English nobility LO religious houses begin to appear. Joan, second wife to Henry IV (from 1403), seems to be the last English queen to jse French in her correspondence. (She was French.)

PROBLEM 4.6: (I) When did English begin to replace French in social usage in

FROV MFDDLE BNQUSR TO HODF.EN ENGTJ5H

77

comparison to French in official usage? (2) Why the dillcrence? (3) Can any of the social patterns be said to be influenced by any of the political or demo^ graphic events V Spoken nJiRlish, French, and Latin on Official Occasions J2.2 It is reported that in 1116, a defendant in court, testifies in English, 12.JO It is reported in 1198 that in court a native English speaker gives evidence through an interpreter, who t ran slates his testimony into French . 14.1 In 1300, a regulation protecting ecclesiastical property is sent to priests, who are ordered to explain it in English to their parishioners. 14.1 In 1301, Edward JI has letters in Latin from the Pope translated into French in order to be read to the army. 143 I г 1327, Edward Ill's presentation of certain privileges to Londoners is made in English. 14.4 In 1337, Edward Ill's claim to the throne of Fry nee is explained to Parliament in English, though the lawyer doing so was proficient in Latin, French, and English. 14.6 In 1356, the oral proceedings of the sheriff's court in London arc ordered to be conducted in English. 14.7 In 1362, oral proceedings in all the courts of England are ordered to be eon dueled in English, in piace of French, which, it is claimed, is little known in the realm, 14.7 In 1363, Parliament is opened in English for the first time. 14.7 In 1363 and 1365, Parliament again is opened in English. 14.8 It has been claimed that French is no longer used in House of Commons debates in the last half of fourteenth century, though the evidence is unclear. 14.9 In 13 81, during the Wat Tjrler rebellion, Richard II addresses people in English. 14.9 In 1381, Parliament is opened with a sermon in English. 14.9 I n 1389, on election day, legal ordinances for the town of Shrewsbury are read to the. citizens in French. 14.10 Proclamations to the citizens of London are made in French until the end of the Fourteenth century. 14,10 In 1394, a formal apology made in the House of Lords is recorded in English, indicating it was spoken in English rather than French, as would have been customary. 14.10 In 1397, a confession by the Duke of Gloucester is conveyed to Parliament in English. 14.1Q In 1399, the deposition of Richard II and the accession of Henry IV is conducted in both English and Latin. Henry claims the throne in a few sentences in English.

78

WOKL>5 AND MEANINGS

15,1 15.1

15.1 15.6 15.9

Henry IV ma к es a short speech i n Engl ish to h is first Par I i a mentThough the record is unclear, debates in the House of Commons in the early fifteenth century seem to be conducted mostly in English, some in French. In 1404, ambassadors to Paris from Henry IV claim, for diplomatic reasons, that they cannot negotiate in French, Judges are still conducting trials in French. IL has been claimed that French was probably discontinued in the House of Lords after 1483.

PROBLEM 4.7: (I) How did the use of written Latin, French, and English in official contexts correspond with the use of spoken Latin, French, and English? (2) How were political events relevant to the use of spoken English?

Spoken English, French, ant) Latin in Social, Educational, and Religions Contexts 11.7

In 1066, William's army is composed of men from every part of France, and of some not from France. 11.8 Odericus Vitalis claims that William tried to learn English, but with no great success. 11.9 In 1085 at the age of 10, Odericus Vitalis, son of an upper-class Norman father and an English mother, is sent from England to Normandy, where, he says, he was like Joseph in Egypt: he heard a language which he did not know. 11.10-During this period there are many citations of abbots, priests, and 12.2 so on, teach i ng and preaching i n English. 12.4 At approximately this time, it is noted that two French canons, ignorant of English, must tell a story to a group of English monks in Latin. 12.4- Henry 11 is perhaps the first English king since William to understand 12.10 English, though he probably does nol speak it. 12.6 John of Salisbury observes that it is fashionable to use French words in English conversation. 116 In the reign of Henry 1Г, it is reported that an English knight retains a Norman to teach his son French, 12.8 Late in this century, a poem complains that although 500 years ago Bede taught and preached in English, it is now nolonger done. 12.8 About 1175, William of Canterbury relates the following story: At a moment of alleged danger, Hclcwisia de Morville warns her husband in English, "Huge de M or vile, ware, ware, ware, Lithulf heth bis swerd adrage (sword drawn)!" 12.9 At about this time, Abbot Samson, a famous churchman, encourages

FROM MIDDLE ENULISH TO MODERN ENGLISH

12.10 12,10 13.1 13.4 13.5 13.6 П,6 13.8 13.9

13.9 13.9 13.10 13.10 14,1

14.3

14.4

79

his monks to preach in French rather than in Latin and better yet in English, as lie is able to do. It is also said that he gave his manor of Thorp to " a certain Englishman " became he was a good farmer and could speak no French. In 1191, in regard to a legal dispute, it is written that one of four supposed knights involved in a legal case cannot speak French. About this time, a young servant is praised "even though" he is country-bred and ignorant of any language except English. In a thirteenth-century romance, a young man is described who spends his time teaching French to ladies in attendance to a countess. RobcrL Grosseteste, Bishop of Lincoln, encourages preaching in English and does so himself. The rules for monks of St. Peter's Westminster forbid the use of English. William of Westminster writes about this time that the common English people despise anyone unable to speak English. A thirteenth-centjry saint's life mentions three French convents where English girls can improve their French. In 1277, a bishop suggests that a letter he has written to a group of nuns be explained to them several times a year in French or English, Walter of Bibbes worth writes his Traite to teach French to children of upper-class families. It imrodutcs French through a description of the everyday objects and actions of common life, the parts of the body, and so on. 11 is reproduced severai times. In a late-thirteenth-century e\emplum there is a story about a steward who is addressed by his lord in French but does not understand, In 12Й4, Bishop Pecham complains that the Fellows at Merton College talk Lnglish at table. By the end of the thirteenth cemury, there are few references to upper-class speakers knowing English. Rules for the Black monks of York order that Latin or French be spoken and forbid Lnglish. About 1300, Robert of Gloucester says that 'Normans could speak only French when they arrived in England, continued to do so and taught their children French. For unless a man knows French, he continues, he is considered of little account. BuL the lower classes, he adds, hold fast to English. In the early 1320'i, the colleges Exeter and Oriel allow undergraduates to speak French conversationally if they cannot speak Latin. English is not expressly forbidden, About 1327, Ranulph Higden in Poljcbronicon says that the English tongue has become corrupted because unlike other countries where children do Lheir lessons in their own language, English children do their lessons in French and upper-class children learn French from

SO

vVOHnS AND MEANINGS

L 4,4

14.9

14.9

14.10

14.10

14.10

15.10

the cradle. He goes on: ". . . and oplontiysch men wo] lyknc hamsylf to gentilmen, and fondej? wip gret bysynes for to speke Freynsch, for lo be more y-tolde of." In 1332, Purliaiaieot decrees that" Lords, barons, knights and worthy men of great towns" should have their children instructed in French so that they would be better able to fight in wars. Sometime before 1380, Oxford University orders that grammar masters make their students construe in both French and English lest French be forgo Lten, In 1385, John Trevisa, commenting on Higden's observation about French and English in J327, says that after the Black Death, instruction in schools switched to English and that French is not so commonly taught to children among the upper classes any more. Consequently children no longer know French. Chaucer's Prioress in The Canterbury Tales is gently mocked for speaking bad French: And Frenssk she spak fid fair с and feiisiy, / After the scale of Stratford site Bowe, j For Frertsh vfParys was to hir unknowe. Though there is some evidence that Richard II is completely bilingual in English and French, Henry IV is generally considered the first monolingual English-speaking monarch. In J396, a new kind of French handbook appears—sets of model conversations designed for commercial travelers, businessmen, and so forth. At the end of the fifteenth century, Caxton says "the rnooste quantyteof the people vnerstonde not latyn ne frensshe here in this noble royameof cnglond."

PROBLEM 4,8: (1) Can any of the examples or anecdotes be considered to reflect political or social changes? (2) Is there a particular period when spoken English becomes important? (3) What evidence is there for class distinctions?

Some Additional Data Relevant lo the Use of English 11.8

12.1 12.5

12.5

Matthew Paris claims that English-Norman marriages were cnctiuraged by William the Conqueror, and, in fact, many Norman noblemen marry English women. In 1100, a Henry 1 document is addressed to all his faithful people, both French and English in Hertfordshire. By the middle of the twelfth century, native Englishmen are giving French names to their children: Humphrey, Stephen, Roger, Robert, Richard, for example. During the reign of Stephen particularly, knighthood becomes

TROM MIDDLE ENGLISH TO MODERN ENGLISH

81

associated with nobility of birth, chivalry, ceremony, and romance through the many branch talcs written in Ibis period. 12.8 Around 1177, in FHalogus dt Scai carlo occurs the observation that, after a century of living together and marrying, it is hardly possible to tell who is of English and who is of Norman descent. 12.9 About Ш З , William FitzStephen writes that among the cities of the world, London is the most renowned, boasting the finest in manners, raiment, and dining. 13.1 Knighthood is beginning to become a burdensome service in the frequent juries of Grand Assizes, tribunals set up to settle property ownership questions and which can be manned only by knights, 14 Through the fourteenth century, knighthood becomes less important as a social class as the increasing wealth of the East Midlands and London create a growing and more affluent commercial class. 15.8 Because of a labor shortage after the Black Death, villenage begins to disappear. Laborers sell their time for money, pay their landlords with money instead of with their labor. 15.Ё In 1476, William Canton introduces printing presses to England, establishing himself in Westminster, just outside London. 15.10 After Carton's death in 14У1, his successor moves the printing business into London. PROBLEM 4.9: Which of these events illuminate or reflect any of the preceding data of linguistic history 7 (7, 10,48, 61,127, 187, 196,216,223,229,241, 242) PROBLEM 4.10: Here are several French borrowings from before and during the Middle English period. (1) What would you have concluded about the nature of English-French contact? (2) Do the kinds of words borrowed reflect anything found in the data presented above? proud, sot, tower, castle, tomber (dancer), market, chancellor.

с 1066-1250 abbot, canon, cardinal, clerk, countess, empress, duke, court, rent, cell, justice, miracle, baptist, datne, prince, chape!y image, lion, reason, pilgrim, saint, virgin, obedience, religion., sermon, prophet, patriarch, archangel, circumcision, sacrament, fruit, sepulchre, custom, admiral, baron, prelate, crown, astronomy, council, nunnery!, abbey, discipline, physician, parishioner, city, crucifix, purgatory, tournament, desert, unicorn, sponge, journey, rob, large, silence, dangerous, jealous, glutton, joy, tempi, witness, chapter, lesson, story, medicine, confessor, constable, heir, chair, galley, butler, canticle, peace, justice, uncle, aunt, cousin, basin, lamp, rose, catch, change, mercy, poor, rich, wait, prove, war, arrive, pay.

$2

WORDS AMD MEANINGS

c J 250-1350 action, cost, deceit, dozen, ease, fault, force, grief, labor, number, opinion, pair, piece, season, sound, square, substance, task, use, bucket, calendar, face, gum, metal, mountain, ocean, people, actual, brief certain, clear, common, contrary, eager, easy, find, hones!, real, second, single, solid, strange, sudden, usual, allow, apply, approach, arrange, carry, close, continue, count, cover, defeat, destroy, excuse, force, form, increase, inform, join, move, please, proceed, push, remember, travel. c. post-1350 adolescence, affability, appellation, cohort, combustion, distribution, harangue, immensity, ingenious, pacification, representation, sumptuous, aggravation, diversity, furtive, prolongation, ravishment, encounter, deraign, sojourn, solace, affray, languor, disparage, conjecture, disdain, explicit, proportion, register, respite. {1,158, 194) 4.1 J: Here is another lists of word*, organized by cultural area. Comment, PROBLEM

a. b. c.
government, state, royal, authority, parliament, assembly, fax, revenue. prince, duke, count, baron, squire, page, sir, madam, peasant, slave. religion, sermon, prayer, clergy, cardinal, chaplain, friar, crucifix, crime, defendant, judge, at tarneyjury, evidence, bail, verdict,fine. army, navy, peace, enemy,battle,combat, defense, soldier, captain. fashion,gown, robe, lace, but ton, boot, sat in, fur, ruby, pearl, blue. dinner, supper, taste, feast, venison, beef, veal, mutton, pork, toast, cream, sugar, salad, lettuce, fruit, cherry, peach, herb, roast, boil, stew, fry, grate, mince, goblet, saucer, plate, plotter, table. h. art, painting, sculpture, music, beauty, color, figure, post, prose, romance, story, tragedy, title, relume, paper, pen, study, logic, grammar. i. palace, mansion, ceiling, chimney, tower, porch, curtain, lamp. j. recreation, leisure, dance, melody, chess, characters, conversation. k, medicine, physician, surgeon, apothecary, malady, pain, plague, stomach. 1. place,part, use, city, large, Sine, state, sure, change, close, course, pay, please, face, quit, coat, brown, air, country, flower, hour, manner, noise, number, people, river, able, large, nice, poor, real, safe, second, carry, move, pass, wait. (]) Pick two or three of these categories {except 1) and add five random words. Are they French borrowings or native? (2) Try specifically to add five native words. (3) Try to replace any set of words above entirely with purely native words. (4) How is category ] significant? 4.12: Albert C. Baugh (7) studied a representative sampling of French words in English to determine the chronology of their entry into Bog&h. He

PROBLEM

FROM MIDDLE ENfiLlSH TO MODERK ENGLISH

8$

compiled the following figures: (1) Before 1100: 2 words, (2) twelfth century: 9 words, (3) thirteenth century: 134- words, (4) 1301-1350: 106 words, (5) 1351-1400: 191? words, (6) fifteenth century: J64 words, (7) sixteenth century: 157 words, (8) seventeenth century: 98 words, (9) eighteenth century: 59 words, [10) nineteenth century: 71 words, (I) How do you explain this distribution? Why was the peak of French imports appearing in print not reached until 300 years after the Invasion? (2) What problem in interpreting these figures is raised by their being derived from written texts ?

NORMAN FRENCH VS. fcNGUSH (10*56-1450) Most of this data speaks for itself, so only a brief resume is needed here. Two dates mark the major turning points in the influence of French on English after 1066. The first is 1204. when King Philip of France seized the Norman estates of barons whose primary allegiance was to King John of England and forced them to choose between France and England. The second date is 134®, -when the Black Death began to sweep across England, accelerating the social changes that had been underway for some years. Before 1204, the dialect of Normandy was naturally enough the prestige dialect of those French in England who maintained close contact with their fellow countrymen in Normandy, traveling back and forth, as they did, some living in both countries. Norman French was very likely spoken: (I) by the vast majority of the French invaders and colonists (the Normans'), but not by all the French who settled in England; (2) relatively soon after 1100 by inost, if not all upper-class native Englishmen; (3) by those mtdd Le-dass commercial Englishmen who had to deal with Normans; and (4) by most of Lhe middlemanagement personnel on the large estates who served French lords and supervised English serfs. Evidence for the use of French among the 90 percent or so lower-class English peasants and laborers is hard to find and evaluate. We can probably dismiss the statements of those who claimed they were writing in AngloFrench, the dialect of Norman French spoken in England, so that both the "lewid" and the learned could understand their romances, "Lewid" or ''common" certainly did not mean illiterate, which by far most Englishmen probably were, but rather those unable to read Latin, the most prestigious language of learning. The vast majority of English speakers, the lower ctass, were undoubtedly monolingual. After a few generations, it is also almost certain that most Normans were bilingual. Those knights and barons on the larger estates and the royalty in their castles were probably able to maintain a French environment in which they required no English, lint the fact that many French knights and

S4

WORDS AND MEANINGS

nobility married English women who taught or had their children taught English suggests that at least some, perhaps most of the upper-class French knew same English. And it is virtually tjertain that those below the level of knights who settled on the land and those businessmen who settled in the cities would have teamed English rather quickly, surrounded as they were by an English-speaking population. Of course, the major unsolved problem is exactly how many Normans settled where. The highest estimate, 200,000 settlers, would mean that by about 1250, one in every eight to twelve on the island was a Frenchman or French-descended, a proportion that would have been considerably higher in the major cities. Since the total population in the major urban centers amounted to less than 75,000 or so, and since we might generously estimate that less than half the Normans who settled in England settled on the land, it would mean that perhaps 100,000 urban Normans would have outnumbered the- English in the major cities. And there is no evidence that Normans took over (he cities to that extent. It is more likely that the French in England originally numbered closer to 20,000 than to 200,000. This would keep the Normans in the minority both on the land and in the cities, a situation that would more logically allow spoken Norman French gradually to disappear after social and commercial ties to the Continent were interrupted in 1204. The example of French in modern Canada is illuminating here. A very high density of speakers is required to keep a language alive over several generations. Ног а century after 1066, written English continued to be used for occasional official purposes after spoken French had widely replaced English as the working language of everyday government and law. in the first century and a half after the Conquest, in fact, written Latin, aot French, replaced English. Not until the thirteenth century did French begin to replace Latin as the official language for charters, deeds, wills, and so on. Probably the most important influence of spoken and written Norman French at this time was that it created a linguistic criterion for upper-middle class membership. Without any question, bilingual Normans searching for a word associated with government, culture, entertainment, and so on would use French words in their English speech. And Englishmen striving to function successfully in an upper-middle class milieu would without any question not only learn some French but imitate the Norman-laced English of bilingual French speakers. The comment by John of Salisbury that it was considered stylish to use French words in English conversations could not be more illuminating. Moreover, within half a century after the Conquest, the native population began giving their children first names like John and Hugh, indicating that the Saxons certainly did not universally despise their Norman upper-class. Thus during this period but particularly from the later twelfth century on, Norman French was strongly influencing the character of uppermiddle class English speech. After 1204 and following the second, smaller political invasion of

FEDM MIDDLE ENGLISH TO MODERN ENGUSH

85

French speakers during ihe reigns of Henry TIT and Edward T, attitudes towards spoken Norman Trench changed considerably. With the cultural center of France now in Paris, Parisian French, not Norman French, became the new prestige dialect. Since the French habitually spoken in England wojfd have been the Norman dialect, itself different from what it was in 1066 and now considered uncouth, pressure to learn "correct" (i.e., Parisian) French would rot be so strongly or so directly felt from immediate conversational contact with upper-middle class Anglo-Norman speakers. Consequently, more and more we find instruction in " g o o d " French a social fact of life. Furthermore, in the fourteenth century, the character of the uppermiddle class was changing. Knighthood was not the romantic profession it had been during the reign of King Stephen. In fact, it was becom ing something of a chore. Moreover, the increasing wealth of the Midlands was creating a commercial class that could command its own kind of respect. This growing wealth seeped down to the laboring class until in the later fourteenth century they rebelled against being tied to the land. They eould profit more by selling their labor to the highest bidder and then paying their lords in money rather than in labor. Thus an important social institution that tended to fix the geographical location of the lower class began to break down. Intensifying this tendency in П48 was the first of a series of catastrophic plagues that swept across England, killing perhaps a third of the population before- the century was out. It was most deadly to the very old. the very young, and to those in the close confines of monasteries. Combined with the social forces already at work, the Black Death made an increasingly fluid social order even more so. Before this, French was still a prestige language, as Higden (S327) pointed out. But after the plague, the schools began to use English instead of French as the language of instruction. And it was at least ordered that the law courts should conduct their business in English (1356, 1362). The order was not obeyed, but the fact that it was given is significant enough. Thus during the period following 1204, spoken Norman French began to fade as a major sociolinguistic influence. Parisian French replaced it, but that was a dialect spoken by the most-hated Frenchmen in England, and one not commonly spoken hy Anglicized Norman families, (Compare attitudes toward the German language during World Wars I and II.) But if during the thirteenth century English was beginning to replace spoken French at almost all levels of society and in many official and unofficial social institutions, written French not only continued to be used well into the fifteenth century, but became even more widespread as the medium for social correspondence. It was not until the late fourteenth century and the first quarter of the fifteenth that the final shift toward English as an official and unofficial national written language occurred. Chaucer wrote The Canterbury Tales in English. Letter-writers began to switch back and forth between French and Fnglish. Official documents and .lecords began to appear in English rather than French. Richard IJ and Henry

86

WOKDS AND MEANTNQS

IV were probably the first English monarchs completely fluent in English. Henry IV, in fad, was probably monolingual. By the middle of the fifteenth century, only the House of Lords and the law courts were still bound to French. The Lords would switch to English in a few decades, but the courts would continue with some of their "law French" until the eighteenth century. (In fact, we can still hear one of the hist archaic remnants of law French when the baililT announces the entry of the judge with Oytt, ayez, ,. . the imperative plural afoir, to hear.) It is also clear from the data that London English was destined to set the national standard. The most prestigious, the richest, the most culturally influential city in the land, London became a melting pot of dialects as Englishmen from all over England migrated to the city. In the fourteenth century, however, one can detect a shift of immigration from the south to immigration from the Midlands, a fact which made the southern dialects somewhat less important at a lime when the national standard was being set. London English as a linguistic standard was irrevocably assured when Caxton set up his printing shop in the London area in the last quarter of the fifteenth century.

КI HI НТП OF CLASSICAL LEARNINC When William invaded England in 1066, be led a politically power, built on a strong feudal base, able to reinforce its power on land with a system of fortifications far superior to the Anglo-Saxons' wooden walls that surrounded their wooden forts. Perched atop earthworks rising many feet above the land, Norman castles easily commanded the surrounding area. The abilities of the Norman architects are amply testified to by the Romanesque cathedrals and castles that today still tower over the English and French countryside. But if the Norman matched the Anglo-Saxon in law and political organization and mastered him in architecture and skills of war, Normans were relative barbarians in scholarship and literature. Under Norman rule, English learning declined, until by the thirteenth century, Paris had replaced England as the center of European intellectual life. The University of Paris was world-famed; French literature was copied, translated, and admired from Norway to Spain to Italy to Jerusalem. Parisian French was not only the language of international chivalry but was challenging Latin as the language оГ international expression. As Paris attracted more scholars, classical learning became more and more widely studied. Latin, of course, exceeded even French in its universality. While by the end of the fifteenth century, it was apparently no great shamt

ГНОМ MIDDLE ENGLISH TO MODERN ENGLISH

87

Гог an Englishman not to know French, a clerk who knew no Latin would not be a clerk-scholar at all. Because French resembled Latin in many ways, it made borrowing from Lalin into French exceptionally easy. Ft is terrain that many of the words English borrowed from French were originally borrowed by French from Latin. In fact, it is often difficult to decide whether a word was borrowed directly from French or directly from Latin. Paradoxically, it was probably the very resurgence of English described above in the fourteenth and fifteenth centuries that led English to adopt, with the thousands of French words it borrowed, thousands of Latin words as well. As English became more and more a language felt to equal French, indeed to be preferable to French in virtually every form of written expression, more and more translators set to translating Latin works into English, as had already been done in French. But as one anonymous writer complained in the early fifteenth century, There ys many wordes in Lai\n that we have na propreEnglyskaccordynge thereto. Having no Anglo-Saxon synonym available and having apparently lost the productive means to generate new compounds for abstract words—the typically Anglo-Saxon way of meeting lexical needs Lute Middle Fnglish translators had to Anglicize Latin words. PROBLEM 4.13: Here are several late Middle English Latin borrowings. What do you conclude about the nature of the contact between Latin and English in this period? abject, adjacent, allegory, conspiracy, contempt, custody, distract,frustrate, genius, gesture, history, homicide, immune, incarnate, include, incredible, incubus, incumbent, index, individual, infancy, inferior, innate, innumerable, intellect, interrupt, juniper, lapidary, legal, limbo, lucrative, lunatic, magnify, niaiefdction, mechanical, minor, missal, moderate, necessary, nercous. notary, ornate, picture, polite, popular, prevent, prirate, project, promote, prosecute, prosody, pidpit,quiet, rational, reject,reprehend, rosary, script, scripture, scrutiny, secular, solar, solitary, spacious, stupor, subdivide, subjugate, submit, subordinate, subscribe, substitute, summary, superabundance, supplicate, suppress, temperate, temporal, testify, tincture, tract, tributary, ulcer, zenith.

EXCI:SS AND REACTION

Because the rebirth of classical learning ultimately ted to a new intellectual tradition in Lngland as vigorous as it was anywhere in the early Renaissance, it is not surprising that the vogue for Latin diction linally reached excessive proportions. Beginning in the late fifteenth century, it became stylish for some

8Я

WORDS AND MEANINGS

to embellish a text with such borrowings as abusion, dispone, diurne, equipoient, palestra!, lenebrough>fecunt8us, reclmatory. Of course, it is probably solely from our modern viewpoint that such words now seem bizarre. During the fifteenth and sixteenth centuries, words such as these and words like camp&uUffus, delineation, dimension, figurative, function, idiom, indignity, penetrate, prolix, and scientific may very well have had the same rhetorical impact. It is only because we have for ore reason or another retained compendious and discarded equipoleiit that equipolcnt is unfamiliar and, from our twentieth century point of view, inflated. As might be- expected, this wholesale borrowing from Latin and French was (and has been) opposed by those who believe that English, after struggling back towards linguistic respectability, was able to express whatever iL had to express in native Lnglish words without the help of foreign intruders. Although a good deal of the opposition was on purely practical grounds— many of the imports were simply opaque " i r k h o r n " terms which were probably unneeded- others objected to such borrowings on principle. Sir John Chcke (1514-57) wrote, toward the end of his life: I am of this opinion that our own tutig shold be written cleane and pure, unmix! and жmangeled with borrowing of other tunges. His attitude is shared by some modern commentators. The author of the most complete, though now outdated, history of the Norman invasion and conquest wrote "This abiding corruption of our language [ believe to have been the one result of the Norman Conquest which has been purely evil'1 (E. A. Freeman, History of the Norman Conquest V. 547, quoted by Baugh). Cheke's position led him to insist that not only should an English term be used wherever possible, but English should make much fuller use of a process of word creation that was lost to serious, scholarly writing when the Normans dominated English literature for so long. Cheke, for example, would write fareseyef for prophet, gainrising for resurrection* leechcraft for medicine, fieshstrings for muscle, iikejamme for parallelogram, and so on. Occasionally, he recommends a single word replacement: mooned for lunatic, hundreder for centttrim, toller for publican, crossed for crucified. But since there are far from enough one-for-one correspondences between AngloSaxon derived words and all the ordinary and learned words from Latin and French, Cheke was really demanding that com pound ing., an area of English grammar that had lost much of its vitality in intellectual prose, become much* more productive and vital than it was. PROBLEM 4.14: Here are several Latinate words we need (or think we need) in English to express Lhe concepts they cover: undiscerniiig, exulting, anxiety, intelligence, proud, patient, magnanimous, spiritless, despondency, precious, reliant, principle, accordant, resolute. They all refer to some aspect of emotion or mental state. But no one piece of these words lints1 them to a single cate-

tttOM MIDDLE EKC11SH TO MODFRN PNGT.TSH

89

gory of meaning, as does i-ogn- in cognition, recognize^ cagftojcenti, cognizant cognizance; or -ceive in receive,perceive, conceive, and deceive. In OL, on (lie other hand, each of the words Anglo-Sa\on writers used to express any one of these ideas began with mod- (which ultimately gives our modem word mood). Here is a list of OE words with their French/Latinderived ModE translations. The first element, mod-, means something approximating what is common to the concepts of mind, spirit, thought, soul. The rough meaning of the second part of the compound is in parentheses: modblind(-blind): undisccrning; modbysgung (-trouble): anxiety; тофН (-fill): proud: modteas (-less): spiritless; modcrxft (-art): intelligence; tnodbh$$iertde (-joyful): exultieg; modge8y!dig (patience): patient; modkasi (least): despondency, want of courage. Here are some literal translations for the second part of some other mod- words. The meaning of the second part of the compound is in parentheses. What do the compounds seem to mean? modrqf (-strength): _ -; modsleSol (-fixed, basic): ; mndteof (-valued): ; modswid (-strong): ; modSrea (-menace): , The capacity to compound has certainly not been lost in English, We coin new ones everyday: moortshot,splashdown, babysit, and so on. Bat by and large, we compound to create words for areas of meaning that are relatively concrete, everyday, common objects and experiences. What we have lost is the OE habit of compounding to create words for more abstract, conceptual areas of experience. PROBLEM 4.15: (I) Make up compound words out of OE stock for five French words in each of the categories 1066-! 250, 1250-1350, and post-1350 in Problem 4.10, For example: sermon; churcfispee
90

WORDS AMD

4.17; What arc the national sources for words in the following semantic fields: obscenities, numbers, body pans, apologies (i.e., Гт sorry, excuse me, pardon me, I apologize, / regret that, . . ) , kitchen utensils (pots, pans, stove, ovett, and so forth) and eating utensils {knife, fork, spoon, saucer, goblet, platter, and so on), astronomical bodies (suij, WOOJJ, star, planet, meteor, cornet, solar system, galaxy, universe, constellation), days of the week and months of the year. Can you explain any pattern of national origin? 4.18: Is there a way (0 account for the pattern of sources for the following words: OE: calf

F : veal

cow

beef

boar swine Jeer chicken sheep

OE: farmer F: tailor; vintner butcher, poulterer woodsman glazier, physician brawn fisherman mason, barber pork, bacon shepherd hunter carpenter, attorney venison poultry skinner painter, spicer chandler, forester mutton miller haberdasher, host baker cook hosier, draper mercer, merchant maid butler, servant, waiter

PROBLLM 4.19: There are several pairs of words in English, one of which we borrowed from Norman French, the other from Central or Parisian trench. (I) Which do you think was borrowed first? (2) Why were both borrowed? (The first word in each pair is Central French, the second Norman French.) chattel-cattle, chase-catch, channel-kernel, lance-latatch, gage-wage, guardward, guarantee-warranty.

Chapter 5 THE ESTABLISHMENT OF MODERN ENGLISH

Once English became generally accepted as the medium for official, literary, and scholarly purposes, and for most upper-class social purposes, only one clement was lacking that would insure its triumph in the English Renaissance —a means to make Linglish texts available to an increasingly literate population. But boots were expensive. Because each one had to be copied manually, the twenty volumes owned by Chaucer's Clerk represented a very targe and expensive library. The missing factor was provided in 1476 when William Caxton (c. 1422МУ I) introduced to England the art of printing. Within 200 years over 20,000 titles in English atone were printed in F.ngland. In the Kith century, one translation after another appeared: Thucydides, Xenophon, Herodotus, Plutarch, Caesar, Livy, Sal lust, Tacitus, Aristotle, Cicero, Seneca, Ovid, Horace, Terence, Theocritus, Homer, St. Augustine, LJoethius, Erasmus, Calvin, Luther. In addition to translations, a vigorous theological literature resulted from the Protestant Reformation and the debates it sparked. Works on history, rhetoric, travel, and so on made the 16th century a rich if generally unappreciated period in the history of English prose. Against those who objected that our language was too impoverished to express grand thoughts, English finally prevailed. The continuing interest in Latin, Gieek, and of course French encouraged 91

92

^ORUB АМГ) MPAKINCS

the constant borrowing of words from those languages. During the sixteenth and seventeenth centuries, we acquired, if we did not retain, such words as atmosphere, dexterity, expectation, appropriate, conspicuous, external, expensive, insane, adapt, benefit, consolidate, exist, mediate, ancephaliie, deruntinate, itiecebrous, exciccate, emacerate, eximious, matssuetude. aspectaWe. suppeditate. Originally Greek words, the following were also borrowed from Latin: caustic, chaos, chronology, critic, dogma, emphasis, enthusiasm, scheme, and system, among many others. In fact, it has been estimated that fully one quarter of the total Latin vocabulary was adopted by English speakers in some form, the majority before the end of the seventeenth century. From about 1450 through 1700, a majority of the approximately 10,000 words borrowed into English were probably Latin. Perhaps half of them have withstood the Lest of time to become permanent additions to the language. (7) The most important social consequence of this wholesale importation is that many of these later Latin borrowings (along with many of those borrowed earlier) have shifted from a vocabulary appropriate only to a. fairly formal style to the kinds of words educated people use in their day-to-day conversation: explain, education, scientific,function, exist, system, disagree, and many more.

CHANGING ATTTTUDES TOWARD ENGLISH

By 1400, it wai clear that English would remain the language of upper-class life and most bureaucratic affairs. Latin and French would continue to be used in the law courts, and the debate would continue through the sixteenth century about the adequacy of English to express refined and philosophical ideas. But from the vantage point of the twentieth century, it is difficult to imagine any other outcome than the emergence of English as a language of intellectual and literary merit. But other attitudes were developing too. As early as the 12th century, Englishmen were commenting on each other's speech, William of Malmesbury (1095-1143), a monk from the south of England, observed that the language of the north, particularly in York, was so crude that Englishmen in the south could not understand it. Chaucer used northern dialect for comedy in the
THE ESTABLISHMENT OF MUDEKN ENGLISH

93

5.1: In the fifteenth, sixteenth, and seventeenth centuries, we find quotes such as the following. What other attitudes toward linguistic- variation were developing? (42) 1.

From William Caxton's 1482 version of John of Trcvisa's ИВ7 translation of Ranulph Higden's Latin description of the state of English in 1327. This particular passage is a combination of Ca\ton 7 Trevisa, and Higden's observations: Therfor it к rhat men of inerdj that ben of myddle englond as it were partners with the endes understands better the side langages northern &. sotnei'n than nor [hern &- southern understands cyther other. AHe the langages of the nort hum ores & speciafly at york is so sharp slytying frotyngand urnshapc that wesothem men may unncth understande that langage I suppose the cause be thai they he nygh to the alyens that speke strauneely. And also by cause that the kynges of englond abydc and dwcllc more in the south countreye than in the north countrey. TJie cause why (hey ahyde more in the south соuntrey than in the north counlrey, is by саны; that ther is hetter come londe more pcple moo noble cytees, & moo prouffyiablc havenes in the south con trey than in the north.

2.

From Sir Thomas Elyot's The Вике Named the Covcrnour (1531): , , , at the lest way . . , speke none enghsshe but that which is cleane polite, perfectly and articulately pronounced, omittinge no lcftrc or siHabie, as rolisshc women often times do of a wanlonnesse, whereby diucrs noble men and gcntilmcnnes chyldren (as I do at this daye knowe) have attained corrupteand foule pronunciation.

3.

From a letter by Henry Dowes, tutor of Thomas Cromwell's (1485-1540) son: He reports that a Mr. Southwell is "dailie h cringe hi me to rcade sumwhat in thenglishe tongue, and advertisenge hinie of the nature!I and true kyndeof pronuntiactin thereof.*'

4.

From John Hart's Method* (1570):".,. the flower of the English tongue is vsud in the Court in London."

5.

From George Puttenhanvs The Arteof Fn^lish Poesy (1539): [The language of the poet should be] naUirall, pure, and the most usuall of all his countrey: and for ihe same purpose rather (hat which is spoken in the kings Court, or in ihc good (ownes and Cities within the land, then in the marches and frontiers, or in port (ownes,,, - neither shall he follow the speach of a crafies man or carter, or other of Ihe inferiour sort, though he be inhabitant or bred in the best towne and Citic in this Kealmc, for such persons doe abuse good speaches by strange accents or ill shapen soundes, and false ortojtraphie. But he shall follow generally the bctler biought up sort, such as the Greckes call iharienies men civil I and graciously hehavoured and bred . . . ye shall thcrfore takt Ihe usuall speach of the Court, and that of Ivondonand the shires lying about London within IK. mylcs, and not much above, ] say this but thai in every shyre of Hngland there be gentlemen

94

WORDS AND MEANINGS

and others that speake but specially write as good Soulherne as we of Middlesex or Surrey do, but not the common people of every shire, to whom (he gentlemen, and also their learned clarkcs do for the most part condescend. 6.

From Alexander Gill's Logimoniia Anglic a (1619; translated by E. J. Dobson.42);

In speech the custom of the learned is the first law. Writing therefore is to be adjusted not to that sound which herdsmen, girls and porters use; but lo that which the learned, or cultivated scholars, use in speaking and recitation. 7. From Charles Butler's English Grammar (1633): He takes his standard from the "Universities and the Citties," 8. From Owen Price's Vocal Organ (1665): l l l have not been guided by our vulgar pronunciation, but by that of London and our Universities, where the language is purely spoken." 9. From Elisha. Colts' Schoolmaster (1674): He bases his standards on the language most in use-among the "generality" of scholars. 10. From an anonymous Right Spelling (1704): London speech is too "fine " and " s m o o t h " ; the speech of the Universities is preferable. Yet another attitude that began to develop was that toward change, Rhetoricans and others commenting on English from the beginning of the seventeenth century began to identify the verbal energy and inventiveness of the first half of their century with unmliness and a lack of decorum, ehange in language with decay and corruption. As English came to be used increasingly to translate the classics and as the primary language for scientific and philosophical writings, many Englishmen became uneasy over the fact that Lhey had nothing like a grammar of Latin to settle questions of usage, no rules to guide those not entirely at ease in the language. A handful of grammars, usually in Latin, appeared from the middle of the sixteenth century, written by those who wanted to provide English with the rules it lacked in comparison to Latin. Spelling reformers also began to examine the orthography of their sixteenth-century English, and at the beginning of the seventeenth century, the first dictionaries began to appear (We shall deal with both the lexicographers and orthoepists more fully in subsequent chapters.) The first explicit, though short-lived attempt to stem what many perceived to be a degeneration of English came in 1664. Two years before, a group of Englishmen of letters and science founded The Royal Society, an institution dedicated to the furthering of scientific interests. In December 1664, it formed a subcommittee to improve the English language. But even though it had the successful examples of the Italian Academy (the Accadcmia della Crusca), and the French Academy (TAc&dHBM Francaise), to follow, the committee met only a few times before disbanding. But the strong belief that the English language had to be purified and

THt ESTABLISHMENT OF MODERN ENGLISH

95

"ascertained." or fived, would not lei the idea of an academy die. In 1697, Daniel Defoe (165У-1731) in his Essay upon Projects proposed a committee of 36 to check the "exorbitance" of writers so that " n o author would have the impudence to coin without their authority.'* In 1712, Jonathan Swift (1667-1745) sent to the Far! of Oxford, the Lord Treasurer of Lingland, his A Proposal fur Correcting, Improving and Ascertaining the English Tongue. I n it, he proposed an academy that would not so much prevent (though he believed that goal was dciiiruble) as control change, admitting carefully screened new words to the language only after defective words had been improved or rejected from the language, shortened words like mob for mobile and extra for extraordinary; words with contracted inflections such as drudg'd, distwh'd, rebuk'd, and Jledg'd, and "low" words such as sham, battler, buHyt shuffling, and palming. Neither proposal achieved its end. But the rational, decorous temper of the eighteenth-century mind continued to insist that the principles of order and decorum apply to language as well as to the arts and philosophy. The English language still lacked a grammar along the lines of neat Latin paradigms; low vulgar words still vied with inflated borrowed ones to corrupt the English vocabulary; speakers and writers still had no standard to appeal to beyond their own uncertain tastes.

Two Authorities Then in 1755, Dr. Samuel Johnson (1709-84), published his Dictionary of tht English Language, a two-volume work that followed over a down earlier dictionaries, but marked the beginning of serious lexicography in English. And seven years later, Robert Lowth (1710-87) published his Short Introduction l« English Grammar, a text that went through at least 22 editions before the nineteenth century and set the tone for attitudes toward language that we still feel loday. Together, they exercised a very great influence on attitudes toward language. As Johnson described it, he "laboured to refine our language to grammatical purity, and to dear it from colloquial barbarisms, licentious idioms, and irregular combinations." He was one of the first to comment explicitly on the level of use of a word, labelling many words as "low," "cant, 4 ' or "ludicrous." Despite the many scholarly defects of his work, it was justlyhailed as a landmark in lexicography. English at last had an index to acceptable and unacceptable words, and a prestigious precedent for making dictionaries the final social arbiter of linguistic usage. (92, 131,153,202,205) A good many grammars had been written before Lowth published his in 1762; some even recognized that the model of Latin grammar was inappropriate to the "genius" of English, that usage by the best writers was a better arbiter of disputed points than artificial rules, no matter how apparently logical they might be. But Lowth's litilc book struck a responsive chord in

96

WORUS AN [J MEANINGS

the temper of the [ale eighteenth century. It provided English with a grammar Ll along Latin lines and, it was hoped, enough advice to teach us to express ourselves with propriety in that Language . . . Besides showing what is right, the matter may be further explained by pointing out what is wrong." In effect, Lowth and those who emulated him set themselves up as ад unincorporated Academy, deciding questions of usage not by observing how educated Englishmen actually spoke and wrote, but by appealing to the past or a kind of rationality that had (and still has) nothing to do with the real nature of language. In the context of the entire structure of the- English language, however, the points over which grammarians quarreled constituted an infinitesimaily small portion ofthat structure. They were questions about the choice between lie and lay, hanged and hung, f had rather or / would rather, between you and I or between you and me, who and whom, It is I or It is me, taller than I or taller than me, different from or different than, the older or oldest of the two, his or him doing that, I will or shall, backwards or back ward, he need not or he needs not, averse from or averse to—the questions numbered fewer than a hundred or so, most of them involving the case of pronouns, verb forms, and the choice of prepositions. The prescriptive rules that were set down illustrate how minor features of language can be seized upon to make large social distinctions. They testify even more to the insecurities of upwardly mobile speakers. Given a finite set of rules about pronoun forms and so on, such speakers can avoid or foliow those rules and know that they are not committing social gaffes that would reveal their changing social status. The advice falls into the same class as that provided by etiquette books, interior decorators, and caterers: When in doubt, rely on institutionalised advice. And equally important, knowledge of rules allows those who art insecure about themselves to judge those others who break the rules. The most common assumption behind this advice was that language should be logical. Occasionally, Latin or the eariier meaning of a word was appealed to, but mtirc often the advice was characterized by the logic of "two negatives make an affirmative." Tn algebra, such might be the case, but as we shall see later, double, triple, even quadruple negatives were the rule in earlier forms of English, and rather than cancelling one another out, they only intensified the sense of negation. Similarly, no historical justification exists for objecting to whose to refer to an inanimate object: Л car whose door wouldn't shut. . . and preferring A car the door of which wouldn't shut. Yet eighteenth, nineteenth, and even some twentieth century prescriptive grammarians would argue that whose should refer only to human referents: A num whose house burned down., . . Their logic defeats their powers of observation. The same reasoning has led many to condemn constructions like more complete or more unique, or double comparatives and sjperlatives such as more wilder or most unkindest. The lo^ic of Latin case led to the choice of

THE ESTABLISHMENT OF MODERN ENGLISH

97

older than I and it isJ. The desire lo regularize what appeared to be random variation led to the codification of' the л/iail -WE'//rule. Some grammarians did indeed resist this kind of logic-chopping. Joseph Priestly (1733—1804), the discoverer of oxygen, wruLe Rudiments of English Grammar in 1761, In it, he forcibly argued that the only criterion to judge usage by was the English of those who used the Language most skillfully. A few grammarians and rhetoricians supported this view, but one need only scan a few modern textbooks to realize what little influence they have had. The same prescriptive rules about the same points (often with many of the same examples) are being set forth today and accepted as gospel just as they were in the eighteenth and nineteenth centuries. In opposition to this prescriptive tone are the modern descriptive linguists (p. 26), who emerged in the early part of this century following a more object]vist, empiricist, behaviodst methodology. Because they made no assumptions about a language before they described it, they made no prior social judgements about the inherent goodness of any particular form that might occur. Because they had to confront languages whose structures bore no resemblance to languages they were familiar with, they rejected the idea that any given language should follow "rational, logical" principles. Thus for a descriptivist, if structures like Can"i nobody tell nobody nothing commonly occur in a dialect, then those structures are grammatical for the speakers in that dialect. If Nobody can tell anybody anything commonly occurs in another dialect, it is grammatical for the speakers in that dialect. Neither structure is more grammatical in itself. They would not disagree Lhal social judgments condemn one form and not the other, at least, the social judgments of those who are most influential in the larger community. But the condemnation, they would insist, must finally rely on conventionalized social judgments, not on any inherent goodness or badness of the grammatical form. Unfortunately, this view is misunderstood by those who are not completely familiar wkh modern linguistic principles. It does not. mean that such linguists "have no standards," that with them, "anything goes." The mistake rests on a failure to distinguish detached, objective observations from hortatory predictions of unfavorable social consequences. And that mistake unquestionably stems from the belief that grammarians and dictionary makers should be social arbiters in the tradition of Bishop Lowth and Dr. Johnson. But a good many social linguists look upon themseives as dispassionate observers and recorders of how people in fact do speak and wriLc in various styles at various social levels and how they feel about how they speak and write. The linguists neither proscribe nor prescribe. The debate Lhal raged over the appearance of Webster's Third New International Dictionary (1961) illustrates this difference. Those who edited W r ^ s Second (1934) accepted the role of linguistic censor by labeling as colloquiai or vulgar, and omitted some words and definitions

98

WORDS AND MEANINGS

altogether because they were not considered to be educated or decent usage. When the average person went to that dictionary, he found Eld vice about how to speak and write so that, in effect, he would not be scorned for linguistic uncouthness. Those who edited the Third, on the other hand, looked upon themselves more as recorders, not prescribe rs. They included in any words not found in the Second, and refrained from marking many others as colloquial or slang. For a great many people, this signalled a shirking of responsibility so heinous that m;my newspapers, libraries, and publishing houses refused to purchase the dictionary and, in riome cases, specifically stipulated that it might not even be referred to. What the controversy attests to is a linguistic insecurity very little different from that recognized by the eighteenth-century grammarians and lexicographers. The difference is that those linguists and grammarians best informed about the nature of language and society today insist that every question about the usage of a particular form must be decided on the basis of actual usage in particular situations and that every answer must be phrased as follows: "Speakers of a particular geographical dialect and socioeconomic level statistically tend to select feature X rather than feature Y in social context A.4' Whether any speaker shun Id select feature X in contest A is not a question to be posed by them, but rather by those who know—or believe they know—whiit the consequences of that choice will be. (201) 5.2 r Many current grammar boots claim that the following constructions are incorrect. On the basis of data you collect from writers and speakers you would consider to be educated, literate, and at least middle class, do you believe that these claims are reliable ? PROBLEM

1. 2. 3. 4t

A sentence should not begin with a coordinating conjunction such as and, but, yet, far, SO, Of, nor. Their should not be used to refer to a singular referent, a$ in Everybody brought their books, Which should refer to a noun, never to a clause: He left, which was too bad is wrong. One should not split infinitives, as in He had intended to secretly enter the meeting which was planning to overthrow the government.

5.3: Dictionaries often use labels like slang, informal. non-standard, casual, indecent, vulgar, and so on to advise a reader about the social status of a word. From that label, it is assumed, he should know how to use the word in the appropriate tin: urns lances, or whether to avoid using the word entirely. Consult the introduction to the following dictionaries to determine what labels they use and how they apply them, and apply those labels as consistently as you can to the list of words provided below. Then look up each word in the various dictionaries to determine how well your judg-

THE ESTABUSHMENr OK MODEHN ENGLISH

99

merits agre* with theirs and, more importantly, how consistently those dictionaries apply their their own standards. Dictionaries: Webster's Third New International Dictionary, Springfield, Mass,, 1961. Webster's Second New International Dictionary. Springfield, Mass., 1934. The American Heritage Dictionary of the English Language, Boston, 19 69, The Oxford EuaJish Dictionary. Oxford, 1933. "] ho Random Housi1 Dictionary of the I\ relish I. angu ape. New York, 1966. Webster's New World Dictionary of the American Language. Cleveland, 1966. Words: sin'i, belch, blue (sad), claptrap, classy, crack (sarcastic remark), cute, egghead, enthuse, dope (narcotic), dumb (uninteWigant), fort, firebug, fop (failure)* fluke (good luck), hugger-mugger, infer (meaning imply), inegardless, jinx, kidQola), kith, knock (to criticize), loaf (not work), lamb (mild mannered person), huh (a drunk), nifty, pretty (as in pretty tired), piss, pipsqueak, root (cheer for), shif, stingy, stinker, to-do, tight (stingy), tight (drunk), wangle, wisecrack. In labelling words in this way, dictionary makers assume that someone uncertain about style and usage will know what the effect of a sentence such as this is: All those pretty dunib eggheads ain't enthused about that. Would such a person who consulted the same dictionary be able to anticipate the rhetorical affects of these sentiences ? We perambulated up ihc street to a restaurant. After checking ourehapeaux, we perused the bill of fare and cogitated about our repast. My friend's visage was puzzled, however, because the appellations of the viands were in Chinese, What does the fact that sttch words as perambulate, chapeau, visage, and so on are not labeled indicate about the assumptions or lack of assumptions on the part of dictionary makers?

BO It ROWINGS FROM OTHER LANGUAGES The rise of a small island nation from its foggy obscurity to the commercial ar id military domination of a large part of the globe meant that, in many

100

WORDS AND MEANINGS

ways, the English speaker came to resemble the Roman in Britain in A.D, 45. In India, in the Near EaSL, in Lhe Pacific, and in the Americas, British businessmen protected by British-Jed soldiers dominated trade and set social standards. In the eighteenth and nineteenth centuries, British wealth allowed England to develop its scientific establishment untJUt was, with France and Germany, the leader of the scientific and industrial world. During the ртте-coionial and colonial periods, English came into contact, «viih a variety of other cultures and languages, PROBLEM 5.4: Here is a sampling of words borrowed from languages we have not yet discussed, grouped by language- and period. Considering both the nature of the words (including style and area of reference) and the periods in which they were borrowed, could you reconstruct the kind of contact we have had with these other languages? The words in italics were borrowed through Krcnch. What does that fact tell you about English relationships with these languages ? Italian Hth-15th centuries: florin, alarm, million, ducat, brigand, bark (a kind of ship), лоту {а kind offish). 16tfi—17th centuries; race., nuncio, artisan, doge, magnifies, mountebank, umbrella, gondoia, carnival, muslachio, attack, rocket, barrack, pistol, cavalier, rnuxkei, squadron, battalion, citadel, bankrupt, contraband, carat, frigate, porcelain, bandit, firm (company), motto, artichoke, cupola, cornice, pedestal, pi a z ra, stucco, portico, grotto, balcony, corridor, catacomb, dado, concert, tmdrigaL, viol da gamba,/u^Mt>, pastel, fresco, sonnet, stanza, canto, caprice, regatta, lagoon, balloon, mx\.s\\n,mercontile} risk, opera, serenade, sonata, spinet, largo, piano, intaglio, profile, vista, cartoon, chiaroscuro, burlesque, ghetto. lSth-l&th centuries: casino, vendetta, mafia, malaria, influenza, bronzet lava, breccia, travertine, mezzanine, colonnade\ arcade, loggia, studio, baroque, figurine, soprano, trombone, viola, cantata, trio, concerto, aria, quartet, finaie. andante, adagio, crescendo, tempo, bravo, piccolo, prima donna, sestet., scherzo, contrapuntal, fiasco,fracas, imbroglio, tirade. Spanish Middle English: cordwain (a leather from Cordova), cork. l<Sth-17th century: peso, cask, real (a coin), anchovy, sherry, spade (on playing cards), galleon, grenade, armada, comrade, tornado, sombrero, cannibal, negro, nigger, mulatto, iguana, alligator, armadillo, sassafras, sarsaparilla, mosquito, banana, cargo, Creole, desperado, matador, tone (the fruit), embargo, parade, guitar, siesta, peon, llama, chinchilla, cockroach, vanilla, barracuda, avocado, barbecue, tortilla, plaza.

THE ESTABLISHMENT OF HODQHN ENGLISH

101

1ЯЙ1-1У1И century: albino, stevedore, quadrille, bolero, flotilla, jade (the minera!), cigar, alpaca, hacienda, porcho, silo, cigarette, guerilla, lasso, mustang, gaucho, rodeo, stampede, lariat, quirt, cinch, bronco, pomp a n c bonanza, pueblo, patio. adobc t scrape, vamoose, canyon, burro, dago, cafeteria, marijuana, vigilante, incommunicado, alfalfa. L w German Lansuages (Dutch, Frisian, Flemish, Plattdeutsch, Afrikaans) Middle English; poll (head), dote, dotard, luff, bounce, snatch, huckster, tackle (as in fishing tackle), boy, booze, wainscot, hobble, splint, kit. 15th-16tfi century: firkin (a small cask), mart, hop (the plant), pickle, spool, rack, sled, excise, buoy, hoist, hose (stockings), bulwark, boor, ]niter, snap, groove, luck, placard. J7th-I8m century: brandy, stoker, smuggle, keelhaul, sloop, cruise, walrus J i b . yawl, knapsack, furlough, blunderbuss, easel, sketeh, stipple, decoy, slur, hanker, snort, snuff, hustle, skate, gin, schooner, pea-jacket, caboose, snow, jn angle, cookie. 19th century: snoop, spook, waffle, boss, dope, aardvark, wildebeest, veldt, commando, trek. High German ltith-17tli century: junker, lobby, carouse, plunder, saber, zinc, sauerkraut, hamster. 18th-19th century: cobalt, shale, quartz, feldspar, gneiss, nickel, meerschaum, pumpernickel, waltz, zig-zag, iceberg, poodle, spilz, dachshund, vermouth, lager, zither, leitmotiv, yodel, protein, paraffin, ohm, poltergeist, rucksack, semester, kindergarten, seminar, noodle, delicatessen, frankfurter, hamburger, pretzel, poker, bum, hex, loafer, livcrwurst, nix. Slavic KSth-17th century: kvass, rouble, czar, steppe, mammoth, knout, ukase, vodka, droshky, samovar, tundra, troika, polka. 20th century: pogrom, soviet, bolshevik, intelligentsia, robot, sputnik, babushka.

Middle Lnglish (almost all acquired through French or Italian): saffron, admiral, mattress, cotton, hazard, camphor, henna, alchemy, alembic, alkali, elixir, zenith, azimuth, almanac, syrup, cipher, sumac, antimony, alcoran, mosque, bedouin, lemon. l
102

WORDS AND MEANINGS

Hebrew Old English (through Latin and Greek): amen, hemp, hosanna, manna, rabbi, Sabbath, Satan, seraphim, cherubim. Middle English (through French): jasper, cinnamon, sappiiire, babel, behemoth, leviathan, cabal, shibboleth, jubilee. Modem Kngliib: hallelujah, shekel, Torah, kosher, kibbutz, Persian Old English (through Latin): pard, tiger, paradise. Middle English (through French); scarlet, roc, checkmate, chess, azure, salamander, taffeta, arsenic, mummy. Modtrn English (usually through other languages);spinach, jasmin, lilac, seersucker, khaki, caravan, bazaar, shawl. Turkic Various times: horde, tulip, vampire, turban, fez, cofiee. Indian Languages (Sanskrit, Hindi, Romany) Old English (through Latin): panther, pepper, ginger. Middle English; sandal 17th-18th century: nabob, guru, pundit, chintz, dungaree, mongoose, punch (the drink), cot, bungalow, tomtom, juggernaut, gunny, bandana, sari, jute, chit, jungle, shampoo, myna. 10th century: thug, puttee, cashmere, pajamas, gazelle, mugger, dumdum, dinghy, loot, polo, chutney, Dravjdian (Tamil, Malay, Tdegu) 16th-18th century: calico, mango, copra, curry, coolie, pariah, atoll, Tibtto-Chmese Old English (through Latin): silk. Modern English: tea, ketchup, kowtow. Japanese 17
THE ESTABLISHMENT OF MODERN ENGLISH

103

AMERICAN ENGLISH

In the last quarter of the eighteenth century, there occurred an event that would diminish and probably will eventually end the prestige of the particular English dialect found in the environs of London: The American colonialists declared and won their independence. The earliest colonialists from England settled, of course, along coastal New England and the Tidewater. Generally, they probably came from the southern and midland dialect areas, from south. and from north and cast of London. Somewhat later, the middle Atlantic states, Pennsylvania around Philadelphia, northern New Jersey and northern Delaware were settled by Englishmen from various regions, though most of the Quakers among them were probably from north midland dialect areas. Then in the middle of the eighteenth century came large numbers of ScotchIrish from Ulster. They had been transplanted earlier by the Crown from Scotland and northern England to Ulster to keep Ireland safe for the monarchy, but fled when they could no longer tolerate British exploitation and friction with the indigenous Irish. Many passed through the earlier coastal settlements to western New England and to the Piedmont of Virginia and the Carolina* to become a buffer against the Indians. Great numbers also settled in Pennsylvania,, along the Susquebunna River and then moved into West Virginia, Tennessee, and Kentucky. Another addition to this eighteenth-century immigration were large numbers of German and Swiss from the Rhineland Palatinate. They kept very much to themselves, however, maintaining their German language and culture in Pennsylvania to become the Pennsylvania Dutch (Deutsch), Other immigrants included French Huguenots, Spanish Jews, Moravians from southeast Germany, and a scattering of other groups that have had very little linguistic influence on our speech. After the settlement of the East, settlers from western New England pushed into upper New York State and then along the edge of the Great Lakes area into Michigan and Wisconsin. Settlers from the South Midlands pushed into Kentucky and then into southern Ohio, most of Indiana and southern Illinois. North Midland settlers later moved into central Ohio, Indiana, and Illinois. Settlement of most of the South and southwest largely originated from the western Carolinas and eastern Tennessee. Later migrations generally followed these patterns across the United States, except that during these migrations, Americans came into contact with thfee other cultures which have contributed words to our language: the Indians, the French in the Mississippi basin, and the Spanish in the Old Southwest. (4,59,110,114, 117, 137, Ш , 145, 173, ISO, 197) 5.5: Here arc some borrowings from American Indian languages. Comment on the kind of contact Americans have had with Indian culture.

104

WORDS AND

Which borrowings from languages discuss4.~d in Problem 5.4 are these Indian borrowings most like? There is another extremely large category of Indian words many of us use every day. What is it? South American (usually through Spanish or Portuguese): canoe, cassava, hammock, hurricane, iguana, potato, maize, petunia, jobacco, condor, chocolate, tomato, chili, avocado, cashew, cougar, ocelot, puma, quinine, coyote,cocoon, North American: racoon, opossum, moccasin, moose, hominy, skunk. muskrat, woodchuck, hickory, tvfetn, caribou, succotash, chipmunk, squash, toboggan, caucus (?)» powwow, nxickinaw, podunk, kayak, chautauqita. PKOGLEM 5.6: Rcexamine the Spanish and High German words borrowed in the eighteenth-twentieth centuries. Speculate which words were borrowed through contact with Spanish and German speakers in this hemisphere and which were borrowed through contact with Spanish and German writing.

During this later settlement, of course, waves of immigrants continued to sweep across the Atlantic. The first large "new" wave began in the early I SCO's when Irish who also could not tolerate British repression or subsist on the meager fertility of a stony [and sailed for (he United SLates by the tens of thousands to settle largely in the cities. Following them were the Germans who bypassed their Pennsylvania Dutch cousins to settle predominantly in the upper Mississippi River basin, in Wisconsin, in Chicago, in St. Louis, in Cincinnati. Along with them came Scandinavians who embraced the land as eagerly as the Irish embraced city life. Settling the upper Mississippi valley, they formed ethnic enclaves as enduring as those of the Germans or Irish, The last great wave of migration began after the Civil War, this time bringing millions from southeastern Europe to settle in the cities along the Eastern seaboard and the midwest, providing the lowest social layer in a fluid but stratified society. Most of these groups have contributed a few words to the general vocabulary of American English, but (outside their own dialect groups) their overall linguistic influence has been negligible. As might be expected, British attitudes toward American English in the last three centuries have ranged from an occasional condescending admiration for its lusty exuberance to a more prevalent contempt for its alleged slovenly corruption of the language of Shakespeare and Milton. These negative attitudes had several causes. First, those most likely to defend a linguistic conservatism were probably among those mnst likely to condemn political radicalism. And since Americans were the radicals of the Western world, any differences in their vocabulary or pronunciation would be condemned as roundly as their politics. Moreover, as we have seen, most British commentators on linguistic

THE ESTABLISHMENT OF MODERN ENULISH

105

propriety condemned any unsanctioned coining of new words or new meanings. Unsanctioncd change was synonymous with corruption. So predictably, the most common specific criticism of American English was that our ancestors gave new meanings to old words and invented new words for old and new meanings. It mattered little that a people making a new world would change their language with the land, that the meanings of words ineviLably had to change as they were calted upon to express new ideas, that new words had to be coined to express meanings—emotional and cognitive—that a citified London standard eould not. And since this emerging American culture was unpolished well into this century, it was inevitable thai new words and new meanings used to describe it would be judged in the same term!;. Indeed, if the English had not condemned American speech, it would have been contrary to all we know about how a group feeling itself culturally superior makes sociolinguistic judgments. Not only our words, but our pronunciation was condemned. And ever this is not surprising, considering some British attitudes towards their own dialects. Although many of Lhe distinctive characteristics of earlier American dialects bad begun to level as the migration westward mingled speakers from all dialect areas of England and America, American English west of eastern New England and the Tidewater area of the Carolina* was in some ways more like northern English than London pronunciation. As early as the twelfth century, northern British dialects were considered by many southern speakers to be barbarously unintelligible at worst, too "broad"' at best. Therefore, once the upper-class educated British visitor penetrated the wilds just west of Boston or Charleston, he would have heard a speech that was probably closer to the English of his North than the English of London. And not unexpectedly, he would condemn it. The earliest American attitudes towards American English predictably split between those who deplored its deviation from educated English usages and those who recognized that American English was gaining a character that reflected the needs of its speakers. Noah Webster {1758-1 &43) wrote in J789: As an independent nation, our honor requires us to have, a system of our own, in language as well as government. Great Britain, whose children we are, and whose language we speat, should no longer be our standard; for the taste of her writers is already corrupted, and her language on the decline. But if it were not so, she is at too great a distance lo be our model, and to instruct us in the principles of our own tongue. from Dissertations on the English Language. But linguistic confidence has not come easily to a people who were long overshadowed by a parent culture with a literary, philosophical, and scientific establishment much superior to their own. If Noah Webster had the confidence to revise English spelling into American spelling (a project we shall investigate later), many of his fellow Americans still suffer from an insecurity that on the

106

WOBI3M AND MEANINGS

one hand encourages high faluttn' talk while on the other it triggers the common man's reaction against it. The nineteenth century and most of the twentieth have seen us search for a linguistic propriety in hoth grammar and vocabulary in ways that reveal our insecurities. One is the plethora of textbooks, correspondence courses, and newspaper columns devoted to teaching the average person how to speak "correctly." Another is a propensity (stretching back into our past) to find new terms for socially embarrassing topics: drumsticks and piano limbs for chicken legs and piano legs, white meat for breast meat, he-cow for bull, powder room for the place that has euphemisms for it stretching back to the sixteenth century. And how many Americans still cringe- just a bit when they have to speak with someone who has what he perceives to be that educated, upper-class, la-de-da British accent? But on the frontier, language, was Less self-conscious. Despite the frequent disapproval of London visitors, our forefathers coined, compounded, and converted words to fit their needs. And when they wanted to play with language, they went back to a tradition of "ink-horn" terms. What could knock out a listener better than telling him about a bodacious and splendiferous lady he once knew who obfuscated and discornbobulated him with her Leetotaciously grandiferous charms? As America filled up with English speakers from a variety of British dialect areas and European countries, and as Americans began moving west, many of the original dialect areas merged until today we have dialects that correspond only slightly to what may have been the original dialects spoken by the first settlers. We can si ill recognize that a lobsterman from Bernard, Maine, speaks differently from a New Yorker, who speaks differently from a Charles to nian, who speaks differently from a Cievelander. And all speak differently from a Mississippi an. But compared with other languages, our dialect differences are relatively slight. (And despite the persistent claim, no one in the hills of Kentucky or West Virginia speaks " pure Elizabethan.14) PROBLEM 5.7: In J92K, under the direction of Hans Kurath, the systematic study of American dialects was begun. Out of it has come a variety of publications: Atwood, E, B. A Suney of Verb Forms In the Eastern United States- Ann Arbor, 1953. Кurath, Hans. Handbook of the Linguistic Geography of New England. Providence, 1939. . A Word Geography of Hit Tastern United States, Ann Arbor, 1949. , and R, ]. McDavid, Jr. The Pronunciation of English in the Atlantic Slates. Ann Arbor, 1961, In coining years, we shall see the publication of more parts of the Linguistic Atlas of the United States and Canada, under the direction of Professor McDavid, and the Dictionary of American Regional English, under the direction of Frederic G. Cassidy.

THE ESTABLISHMENT OF MODERN ENGLISH

167

Here are a variety of terms used in various parts of the United States. Survey students frum a variety of areas to determine which terms they have used от heard used. Then try to map the areas of usage 12. 3. 4. 5. 6. I, H. 9. 10. II,

curds о г curd cheese, crvds or crudcheese, sour tnitk cheese, clabber cheese) Dutch cheese, pot cheese, smearcase, cottage cheese. darning needle, snake feeder, snake doctor, mosquito hawk, dragon fly, pail, bucket. sick in the stomach, sick at the stomach. quarter of, quarter tit}, quarter to (as ill quarter to six). belly-bumper, beUy-sfarntner, beSy-flopper, beUy-buster. teeter-totter, see-saw, teeter-board. earthworm, angleworm,fishingworm, fish bait, redworm. frying pan, skillet, spider. snap beans, siring beans, green beansbaby carriage, baby buggy, baby coach, baby cab-

Black English There remains a very large movement of peoples across the Atlantic to this continent that must he included in any survey of linguistic influences on American English. Perhaps more than any other n on-English group they have influenced our patterns of life. And perhaps more than any other group, they have preserved certain of the earliest influences in their present speech. The sordid history of the slave trade need not be rehearsed here in detail. Slavery has afflicted virtjaily every civtiination, though firearms and sailing ships made it particularly easy for Europeans to practice it for very large profits. The first slaves merchandized in the Colonies were sold to the Jamestown seLllcrs in 1619 by a passing Dutch privateer, who had probably stolen them from some Spanish or Portuguese trader, who had probably purchased them from one of the siave ports along the West Coast of Africa, now the countries of Ghana, Ivory Coast, Liberia, Sierra Leone, and Guinea, where prisons were built t» store slaves until they could be sold to European merchantmen, usually Portuguese. The word Negro was borrowed from Portuguese or Spanish, nigger from neger, a Middle French word also borrowed from Spanish or Portuguese. Both words ultimately can be traced back to Latin niger (black). At about this time, there arose among the European traders and their African counterparts pidgin languages, mixtures of African and European words in a simplified grammatical structure but one still complex enough to prevent us from thinking of them as a childlike, primitive form of communication of the " N o ticlcee, no washee" stereotype. It is very likely that at least some of the slaves learned one of the pidgins, for the diversity of languages

10$

TVOftl'JS AND MKANINGS

they spoke in captivity probably would have made communication difficult enough to require some lingua franca to enable overseers to give the simplest orders. (40,77) When these slaves were herded into the loathsome holds of the slave ships, they became a highly perishable commodity, so the slavers were eager to sell them before they succumbed со starvation and disease. The nearest port in the WcsLcrn Hemisphere at which a profit could be turned on their cargo was in the West Indies, where the slavers traded their passengers for sugar, rum, and other West Indian produce. The next closest North American port was Charleston, South Carolina, It became the principal American point of entry for both West Indian and African slaves. Since the economic problems of carrying them once more up the coast were generally greater than finding a profitable way to work them in the area where they first arrived, it was probably geographical and economic eons [derations, not lower moral standards, that eventually concentrated slavery in the South. Thrown together as at first they were into very large groups, it is likely that one of the original Atlantic pidgins became the basis for a pidgin used by the first slaves in this country. The earliest literary representations of slave dialect rese-mble them: Kay, massa, (says he), you must leave me, me sit here, great fish jump into de canoe, here he be, massa, fine fish, mass. Kroni J. F, D. Smyth, A Tour of the United Stales of America, Воссагигта make de Black Man workee, make de Horse workee, make de Ом workee, make cbcry thing workec. Quoted by Benjamin Franklin, "Information (o Those Who Would Remove to America." (both quoted in (39)) The use of the object form rue as a subject, the lack of inflections, the undilTcrcntiated be, the enclitic -ее on verbs and other parts of speech (as in the No tickee no washes cliche)—all are characteristics of the pidgins developed in the Atlantic area. As we shall see in subsequent chapters, some linguists have argued that certain features of Modern Black English may be descended from those earlier features. (39, \Ъ1) One of two things happens to a pidgin: It either disappears, leaving a few traces behind, or, if it becomes the working language for a large enough group of speakers and develops a grammatical structure and vocabulary complex enough for extended communication, it becomes a creolizcd language. While the first alternative was probably the rule in this country, the second occurred in the restricted area in the islands off the Carolinas. A Creole language called Gullah developed which preserved not only many grammatical features of pidgin but many words from. African languages as

THE ESTABLISHMENT OP MODERN ENGLISH

109

well. Some words of undoubted African origin which standard English has borrowed include goober, jazz, hoodoajioodoo, tat?, gutnbo, Ьащо, chigger, juke, and ok in. (224) Now, a major controversy in modern dialectology is whether this view is, in fact, correct. Can we trace the features of any dialect of Black English (excluding Ciullah) back to its pidgin origins, or are most modern features, if not all, actually characteristics of early Modern English learned from white English overseers and retained in the Black community because of its social isolation? We shall examine the evidence for both claims in subsequent chapters

American English: Л Summary If the effect of America and the New World on Modern English has been socially significant, grammatically it has been very slight. The question of pronunciation we shall deal with later. The principal lexical consequence has been a relative handful of words {excluding place names) borrowed from the Indians, a good many new meanings attached to old words, and an abundance of new compound words: space age, cocktail flow, coffee break, and so on. But in comparison with the over half a million words in the language and the multi-million meanings attached to those words, the lexical differences between American and British English are minor, and while the number of new meanings and compounds coined ал this side of the Atlantic is not inconsiderable, they arc still a small minority in the total lexicon, and not a significant part of the central core vocabulary. Perhaps the most interesting difference between American and British English is the modest degree to which American English is not more "advanced " or "modern" than British English, but occasionally more conservative, both in pronunciation and lexicon. АЦ these nest words and meanings, for example, were once common in British English. They were transported to the United States, where their form and general meaning were preserved. But they either dropped out of British English or changed to a new meaning. Thus American English often has an older, more traditional meaning for a word than doe* British English: loan (as a verb), progress (as a verb), bug (as a general word for insect), druggist, wilt, sick (as a genera! word for ill), apartment, tariff (us the narrow meaning of tax on imports), baggage, raise (as in raises vegetables), quit (as in quit work), 5.S: After tracing as many words as we can back through OE to their Germanic and Indo-European sources or through the various other languages if they were borrowed, there remains a residue of words which the Oxford English Dictionary lists as of "unknown* 1 or "obscure" origins. Here is a sampling. Comment.

110

WORDS AND

clasp, clog, cobbler, cub., dandle, dodge, dumps {as down in the dumps), gadget, hug,joh7 nod, kill, paddle, pang, pedlar, pelt (throw), pimpk, lad, pick [verb), pie, pink, plot, plump, prod, primp, prong, puke, pun, punk, pussy, puzzle, queer, quirk, quiz* ramble, rig, rip, roam, rote, rove, row (fight), ruffle, rust, scads, scatter, scoop, scorch, scoundrel, scowl, sedan, shabby, shack, sham, shoddy, shrug, skid, skull, slab, slung, slog, slouch, slum, slush, slut, smug, snatch, sneak, snide, snob, smug, sprain, spree, spry, squander, squid, struggle, stroll, suds, sulk, swizzle, tarry, taunt, tic, tickle, tiddly wink, tiff, tip, toad, toast, trash, trim (neat), trudge, twaddle, tweek, wad, wisp, yank.

SOME CONSEQUENCES OF A COSMOPOLITAN VOCABULARY

Word Families One fascinating consequence of this large-scale borrowing from other languages into English is that over the years, we have built up clusters of etymologically related words that may differ considerably in their modern meaning and form, but which all go back to the same root. All these words, for example, have as their uftimate root the Indo-European *pod-, or foot: foot, fetter, pedigree, pew, pilot, pedestal, peon, pawn, pedal, pedestrian, podium, expedient, expedite, expedition, impediment, impeach, repudiate, octopus, trapeze, trapezoid, tripod, sesquipedalian, podiatrist, cephalopoda Among other Forms the IE root split into a Germanic *fot-, a Latin *ped-, and a Greek *pod-. From that point the words began to change in form and meaning, migrating from one language to another, picking up new senses from each culture. We can mughiy represent some of these relationships on a chart like Figure 5-L 5.9: Here are some other words based on the Indo-European stem *pod-. Trace their descent, fetlock, pawn (as in chess), octopus, antipodes, tripod, impeach, expedient, expedite, impediment, podiatrist, cephalapod. Do the languages from which these words immediately come correspond to particular centuries? PROBLEM

5.10: Here are groups of words which in part or as a whole can be traced to a common root. Pick one group and, referring to Eric Partridge's Origins, plot the derivations of the words, using the model in Figure 5,1. PROBLEM

1 - amateur, enemy, amorous, amiable. 2, camera, chum, chamber, comrade.

Indo-European "podI Germanic *fot-

OE

Latin *ped-

Greek *pod-

I

French

Italian

Spanish

Medieval/ Late Latin

A. OE foot B. ОЕ fetter С

1440 pedigree *— pie-de-grue

D. 1475 pew *

puye <

E- 1530 pilot *—

pilot ^

podia pilota <

podion pfrda

F- 1563 pedestal 1 — piedestal <— piedeatallo G. 1609 peon < H, 1611 pedal *

peon pedale ^ -

pedale 4 -

ped orient pedale

L 1716 pedestrian *

pedestri

J. [789 podium •<

podium *• Figure 5.1

A. Foot comes directly from (he Indo-European through Germanic. B. Fetter, to bind the feet, also derives through Germanic. С Pedigree has a curious hislory. In Lhe genealogical records of medieval France, a symbol of three lines, allegedly resembling a crane's foot, was used to show "descended from." This was tailed a piv-ite-grtte, or foot of the crane. D. Pew. Greek /ют has a genitive podos, which has a diminutive, podion, a base or pedestal. The Laic Latin plural, podia, was apparently understood as a singular in Old French and became puie, a raised place. In Middle French this became puye, a parapet or balcony. This became in Middle English ptwr tben pevc, then pew, a preacher's raised stall. E. Pifor Greek potts, whose oblique stem was pod- gives pedon, an oar blade, whose plural was peda, a rudder. From this came the Byzantine Greek *pido/es> giving Italian pedota, then pilot a. Medieval French borrowed it as pilot, then early Modern French, pilote, the source for English pilot, F. Pedestal. From Italian piede di math, for foot of stall: support for a stall for standing or sitting, becoming in early Modern French piidestal, G. Peon. Latin pis, foot, ftives Late Lai in fc. A.P. 200-600) pedo, a foot soldier. The ohlique stem 'upeden, which in Spanish becameрчдиH. Pedal. The Latin pedale became in Italian pedaie, a footstool. Early Modern French borrows it ач pedal?, and English borrows it as pedal, originally a treadle on a musical jnslruinent, I- Pedestrian. The Latin pedester with its oblique stem pedestri- means of the feet, or lowly. English horrows indirectly from Latin and adds the -inn suffe. Us first meaning was I he 'L io^Ly, humble" sense, as in "pedestrian prose." Only later did it come to mean someone who could not afford to ride; i.e., someone who must walk in the street. J. Podium. Л «borrowing of the root for pew. The Greek diminutive podlon gives Late Latin podium, which was borrowed directly into English. (161) 111

112

WORDS AMD MEANINGS

3. 4. 5. 6. 7. 8. 9. Ml, 11. 12. 13. 14. 15. 16. 17. 18. 19.

cream, Christ, chrism, grind, cretin. Diana, deity, diary, dismal,fturaal,jovial, Tuesday. father, repair,perpetrate,patriot,pope. genera!, genitals, genius, gen tteman, engineer. joinjun t a, jostle, yoga, yoke, zygoie, Ugh t, lucid, hma, luxury, leukemia. 6M», forlorn, analyze, pal$y, solve, irresolute. manual, manner,maneuver, manure, manuscript, commando, demand. measure, menstruate, dimension, moon, metric, parameter, month, Monday. nerve, needle, enervate, neurotic, a, ane, alone, none, any, atone, eleven, union, universet ounce, inch. ply, pliers, apply, explicate, display, employ, supplicate, simple. sport, portfolio, export, rapport, support. punt,pungent,punctual,point,pugilist,pygmy,pugnacious. rapt, rape, rapture, raven, ravine, harpy. sit, settle, sedation, size, dissident .obsess, president, reside. stand, static, stage, ecstacy, prostrate, system, systaltic, stanza, statue, state,statistic, staid,siandard, varlet, vassal 20. thing, thumb, thimble, tomb, tumescent, tuberculosis,protrude, truffle.

Semantic Families

There is another consequence of this wholesale borrowing. English perhaps more than any other language has clusters of words that, while very different in form and not precisely synonymous, are very similar in meaning. In most cases, we have at leasL pairs of words, one native, the other usually a French or Latin import: hearty-cordiaL help-aid, need-desire, buy-purchase, die-expire, climb-ascend, eat-dine, need-require, work-labor, teach-instruct. This has made it possible for English speakers more than the speakers of any other language to select words with fine nuances of meaning and association . Larger clusters of words like help, aid, assist, succour, relieve, reinforce, support, and so on divide the general semantic field they cover into much finer divisions than do the vocabularies of other languages. The very richness of our vocabulary, however, makes choosing the right word more complicated than in other languages because it also makes wrong choices possible. One often reads the advice that "all things being equal," a writer can avoid the problem by choosing the Anglo-Saxon word. But this is often bad advice. These words, for example, help-aid-assist, do not mean the same thing. Each covers a slightly different range of meaning. This wide-scale borrowing has created a vocabulary that allows an English speaker to modulate his tone, to control the formality or informality of his language to fit the needs of his rhetorical situation, or by using cerf a in. words, create the rhetorical situations he wants. And conversely, it allows

THE ESTABLISHMENT OF MODERN ENGLISH

113

him to make social mistakes. It is not difficult, for example, to choose which of these is more appropriate to a formal situation: 1. 2.

The requirement that individuals present must proffer identification an request is hereby instituted. From now on, everybody around here will have to show who they are if somebody asks (hem.

The need for such stylistic possibilities, of course, results from the range of social situations we find ourselves in. The problem for most speakers is to modulate their style to fit each situation with its unique social variables. Most native speakers know how to behave in the context of their own class in the situations in which they normally find themselves. But when they move out of their class, up or down, they enter another social milieu whose speakers may use different signals, or the same signals more often or less often than in another class. In general, a lower-middle class speaker aspiring to a higher social class and sensitive to the verbal cues in that higher class is fikety to overcompensate. He will use too many upper-middle class verbal cues when he finds himself in a fairly formal situation in which he is self-conscious about his language. Thus, though we have enriched the language with clusters of words that allow us to distinguish very exactly fine shades of meaning, tone, and style, we have created a complex instrument that, used badly, can have unfortunate rhetorical consequences. Besides their inappropriate social usage, French and Latin borrowings are also used by bad writers of ModE to produce a fog of verbiage that obscures thought. Here is a passage, thick with Latin isms, from a professional journal: Withdrawing positive reinforcement and eliminating the opportunity to obtain positive reinforcement arc commonly employed in our society as punishments for \tth&v\oi judged Xo be undesirable. Sentence to prison^ dismissal from work, Joss of driver's license, and withheld allowances are a few examples. It is only in the last decade, however, that the aversbe properties of the removal of positive reinforcement have begun to receive an experimental analysis of some depth, (28 Romance words out of 70). The writer could have said the same thing more clearly in fewer words: We often punish those who behave badly by taking away things we give them to he-have well. We lock them up, me them, take away tbeirdd ver's ficenxe., and witlihold their allowances. But only in the last (en years has anyone experimented in. depth to find out whaf happens when we do this. (4 Romancewords out of 53), The author was writing in what he perceived to be a very formal contest on an important subject—a rhetorical situation that required a very formal

114

WORDS AND MEANINGS

rhetorical style. Unfortunately, as happens so often, he sacrificed clarity and vigor for the formal jargon of his profession, a jargon strongly depend en L on French and Latin borrowings, and on the habit of turning verbs and adjectives into nouns: withdraw-wilhdrawing, reinforce-rewfarcement, elimsnateelardnaiing, punish-punishment, behare-behavior, sentence-sentence, dismissdismissal, lose-loss, allow-altowance, remove—removal, analyze-anaiysis, deepdepth. fROBLFM 5,11: Here are groups of rough synonyms. Can you distinguish more than two levels of style among these? Make up several sentences that mean roughly the same thing, but consistently use words from different stylistic levels. For example; They ripped off some chintzy threadsjThey purloined some inexpensive attire [They stole, some cheap cfotttes. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. IS. 19. 20. 21. 22.

abduct, kidnap,snatch. account, statement, bill, tab. acknowledge, confess, admit,ownup, sing. addict, fiend, fan, head. deceive, betray, double cross. delay, procrastinate, dawdle, dillydally. purloin,pilfer, steal, swipe, rip off, cop, lift. distnisSifire, bounce, discharge, sack. dupe, trick, hoodwink ,fool. face, visage, countenance, mug,puss. reprove, rebuke, reprimand, chide, bawl out. chicken oiti,g:i-e up, capitulate, submit, surrender. theft, larceny, burglary, robbery, rip off. thin, skinny, sknder, dim, tenuous. guy, gentleman, person, man ,felto w, chap, rags, clothes, attire, threads. flicks Jilms, movies, cinema, motion pictures. food, grub, mauds. intelligent, smart, bright, sharp. jock,athlete,sportsman, dumb, stupid, dopey, unintelligent. confused, puzzled, screwed up, troubled.

A final possibly unfortunate consequence of these foreign imports from Latin and French is more speculative and certainly more controversial. 5.12: Here are lists of Latin/French and native words. Read each list to another person, asking him to give the native synonym for the borrowed word in list (1) and then the borrowed synonym for the native word in list (2). Keep track of cxattly how many seconds It takes to complete each list. PROBLEM

THE ESTABLISHMENT OF MODtRN ENGLISH

US

1.

T^tin/French Words: repair, excavate, depart, velocity, rapid, decay, illumination, at tempi, terminate, converse, desist, injure, prevaricate, rotate, intense.

I.

Native Words: chew, eat, sell, deadly, answer, lengthwise, watch, think, sad, believable., ask, tell, raise, old, walk.

Unless the person has an unusually ready vocabulary, it probably took significantly longer to think of the corresponding words for the native list than for the borrowed list. What consequences does this have for recognizing, and for using the vocabulary of formal English? What social consequences are there for learning or not learning the vocabulary of relatively formal English?

For the most part, serious writing—the writing of government, of much business and commerce, оГeducation, and so on—relies on French and Latin imports over native words for content words, words that convey ideas rather than grammatical meanings, as do the native function words such as in, thct may, to, of, with, will, be, though, while, for, and so on (as the formal passage preceding Problem 5.11 indicates). This means that if a native speaker wants to take part in the intellectual, academic, or governmental life of our society, a life often characterized by formal language, he must learn a considerably larger vocabulary ihan speakers of many other languages. Consequently, the most serious criticism we can mate of the F.nglish vocabulary is that which the Danish linguist Otto Jespersen made: English is an "undemocratic" language. Those who want to use English for more than casual communication have to learn a very long list of separate items. Contrast this with the list of words in Problem 4.13. Each word that roughJy corresponds to a Latin or French word is made up of parts semantically transparent. This means that the OE speaker had to learn many fewer individual words. He had to learn only a basic core of words and the grammatical rules for compounding and understanding them, a tacit knowledge presumably shared by every native Anglo-Saxon speaker. This does not suggest that we should have no formal linguistic cues for formal social situations. Levels of formality in social behavior, particularly in regard LO language, are an aspect of human behavior so widespread as to be very possibly a linguistic universal among even moderately complex societies. All human languages seem to have verbal cues to indicate when something serious is being said in a serious contest. It just so happens that among a great many other cues, English has selected as a very important one a French and Latin vocabulary that has to be learned virtually word by word, . PftOBj,EM 5.13: There is a kind of borrowing not yet mentioned. The words

116

WORDS AND MIZAN-|NG5

are called loan translations. Rather than borrowing the words themselves, we borrow the meaning of the word or phrase and then translate it into English, Here are some examples. From French: it goes without saying, gilded youth, castles in Spain, >•> [arriage of convenience, 1гЫ balloon. From German: superman, academic freedom, beer garden, world view, wonder child. In other cases, we borrow the words themselves and change them very little: lebensraum (room for living), Zeitgeist (spirit of the times), realpolitik (politics of realism), blitzkrieg (lightning war), weltschmerz (pain of the world), /юг.-? d'oeuvre, laissez-faire, sat-oirfaiw, aide-de-camp, raison d'etre, tete-d-tete, pie-a-la-mode. Why do we on one occasion borrow a word without anglicizing it: Zeitgeist and raison d'etre; on another borrow a word and anglicize it: garage, menu, ration; and on yet another, borrow the meaning without the words: academicfreedom and marriage of convenience'? At this point, it might be worth distinguishing between two large categories of borrowed words. On the one hand, we borrow words to name objects uniquely associated with the culture in which the objects are found; florin, peso, pickle, paraffin, samovar, mosque, kibbutz, chess, fez, mongoose, mange, ketchup, geisha, cockatoo, penguin, cougar, skank. In some cases, the object becomes entirely absorbed into our culture: pickle, paraffin, chess, ketchup, cougar,skunk. In others, the object remains identified with that other culture: florin, peso, samovar, mosque, kibbutz, fez, mongoose, mange, geisha. While it h true that these latter words are certainly part of the English vocabulary, they are used only to refer to objects still part of that foreign culture. On the other hand, some borrowed words seem to tap concepts in our culture that, until we borrowed the words, had no names, or at least no names that captured the distinctive quality of those concepts: blarney, slogan, creed, guess, snub, sly, inform, podtmk, ghetto, vigilante, luck, intelligentsia, zero, kosher, bazaar, horde, pundit, kowtow, tycoon. For one reason or another, these words widened their sphere of reference to include concepts and experiences in our own culture. These words now cover "psychoeultural" aspects of the English-speaking world. It would be very difficult to imagine how their referents could disappear. Indeed, it is difficult to imagine how we got along without such words before we borrowed them. We certainly had other words that probably included the semantic area of these words, but no OF word that exactly represented the precise semantic reference. In fact, it may be that the word, as it were, crystallized or precipitated a semantic focus where previously there was only an amorphous sense of something in the culture that was never spoken of in precisely the right way.

THE ESTABLISHMENT OF MODERN JiNOLISH

117

Such a concept is difficult to prove, but the fact remains that in many instances, we seize on borrowed words not for what they refer to in their native culture but what they allow us to refer to in our own. 5.14: In the lists of borrowed words provided so far, pick out ten words that seem to have no satisfactory synonym in. our native vocabulary, (Be aware of the fact that there may at one time have been such a word in OE orME, but that it is now lost.) PROBLEM

Chapter 6 CREATING NEW WORDS

So far, we have examined two sources of our current vocabulary: our native inheritance, those words we cannot identify as being borrowed, and borrowing, the most productive source of completely new forms. In other languages, German for example, borrowing is considerably less important as a source of new words. Like the Anglo-Saxons, Germans make up compound words for new ideas out of their native stock. While we borrow oxygtn from the French, the Germans synthesize saue rsrt>ff{$om + stuff) out of two words they already have. But if English h;i5 borrowed hundreds of thousands of words, yet we actively make up as many more new words out of what is available to us in our working vocabulary and patterns of sounds. There are three large categories of such word formation. In varying degrees and in different ways, all three are rutc-fiovcrncd. That is, we do not make up new words capriciously, but according to rather specific constraints lhat we induce as we learn English. The first large process of word formation, which includes compounding and derivation, is very strongly rule-governed. To create new words in this way, a speaker has to know both individual words and pieces of words, called affixes, and the grammatical rules which tell him how to combine words or attach affixes tu words, and how to change the pronunciation of both the root

119

120

WORDS AND MEANINGS

(the base word Lo which the affixes attaches, as in decide-decteion) and the affix (compare divert-diversion and insert-insertion). The second process of word formation is also rule-governed, but the rules are psychological rather than grammatical. There is in the human psyche something we can call the principle of least effort, or a "simplicity" principle. It leads to new words for old meanings and some new meanings for old words by shortening a word or phrase: dorr» for dormitory, pro for professivnui, taxi for taxi-meter cabriolet^ jet Гот jet-propelled aircraft. Also included arc blends like smog from smoke and fog; acronyms, words made from the first letters of a ptuase or title; NATO, NOW, HOPE, and so on. The third process may also be rule governed, bat the rules, also psychological, are much more difficult to formulate. These are the rules which direct us in creating new roots, completely new words that did not exist before someone, consciously or unconsciously, made them up. Included are root creations, words simply concocted out of thin air like Kodak or nylon; and echoic and symbolic words, words which allegedly sound like the thing they refer to; pop, crackle, clang; teeny, grump, spit.

TWO GRAMMATICAL PROCESSES Com pounding As we have seen, one of the richest resources of OE, and a source that many have said lias greatly weakened since then, is compounding—putting together two independent words to make one word. It is still a common process: Green and home go together to make greenhouse, as opposed to a green house. 6.1: Using these ne*t examples, suggest some criteria for distinguishing compound words {comic.strip) from ordinary phrases (a comic strip). What criterion can not be used? post efftee, grammar school, boyfriend, coal mine, gym teacher, rifle- range, horse-fly, pot-pie, water-cbek, policeman, airport, bloodhound, bedtime, horsewhip, bagpipe. PftOBLGM 6.2: All the compounds in Problem 6.1 are of the form noun +• noun = iwun. Tliat is, a noun, horse, is added to another noun tfiy, to make a newnoun, horsefly. But if we use all combinations of parts of speech and different verb forms: infinitive + noun = pickpocket; V-ing + noun = flying machine; and V-td I поил = ground meat, we could, theoretically, create well over 150 combinations. In Table 6.J, the vertical column at the icft indicates the

CREATING NEW WORDS

121

possible first wurd in the combination. The horizontal row indicates the part of speech for the second word in the combination. The vertical column at the far right indicates the part of speech that results from the conibination. The entry for noun + noun = noun could be filled in with: spot + light = spotlight; the adjective -I- verrwn^ combination with: easy + going; and the preposition + infinitive combination with: over + see. Use Table f>.l to inventory all the possible combinations and to answer these questions: 1. 2.

3.

4.

5.

6.

Which compounds are the most pnHluctivc today? That is, in which categories can we freely coin new compounds, not just find examples? Which compound* are impossible in ModE? Which parts of speech are most productively involved in compounding? The major ОЫ compound types arc entered in their appropriate boxes with translations. Comment, A count of the first 500 words under F in Hall's Concise Anglo-Saxon Dictionary reveals about 40 percent compound words. Count the first 5Ш words in any standard desk dictionary under two or three different letters. What do you conclude ? In a study of compounds in running lines of OH. poetry, С. Т. Сагг counted a compound about every two lines, (26) Assuming about six words to a line, that comes to about one compound every twelve words. How docs this compare with Mod В poetry'.' ModL prose ? Even where the language seems Lo resist it, make up a new compound word. That is, verb +- verb = verb does not seem natural in English, but we can imagine it: to sneer-smile. When you have created new types, determine whether the meaning you would assign them сип be inferred by someone who hears them for the first time. Does the problem of how to define a part of speech complicate this exercise? Are there any questionable categories in Table 6.1 ?

We might conclude from this very brief inquiry that compounding is less productive than it was in OE. There neither arc as many individual compound words in the vocabulary nor do they seem to occur as frequently. Sometime between the Invasion and the end of Latin and Norman French as the intellectual languages of England, English speakers stopped casually compounding words for abstract scholatly ideas and began borrowing words for them. That compounding is far from dead in certain styles and contexts, though, is apparent from our modern compounds and our readiness to create new ones: Anchorman Walter Cronkite has learned that mission control is announcing a holdup in the counftiowii for today's tftoonshof. The backup crew at the is on standby. A follow-up story will continue after the station

TABLE*.!

CREATING COMPOUND WUROS -Й1Й

„™,

boi-craft (huDlcrnft) doiB-pwr ISEBd I-..I I',

noun

vriu

NOUN

(pjud НиГЦ hjcalT-Jitpendc лШБСПУЕ

(ttad-t|4rir)

qcir-rvijcleO

UV

• ul-: :» .-i iv,

(viEe-mlkdedt

"US

*." INF

V...

iivretmaji

M:I

.N,

-.

-Eb

INF

VEK»

- ^

JiDJ

-ED

W

-1Л

VfP»

-Eb

INF

vrr»

-IV&

*

-ED

NW-'N

(in-gaing)*

(uvcrLhrawri)

и)чтз.Ы

pumerraw

fifcf-tlnf^n (uvirdTi""e) + ЛцГ|| 1.1

CHEATING NT-W WORDS

123

In fad, the claim that compounding has weakened since OF. times may be true for only certain genres of literature. If we look at OE poetry, compounds do indeed abound. But OE prose has many fewer (though StiD more than ModE prose). If we examine ModE prose- from editorial pages, textbooks, and not self-consciously casual magazine articles about moderately abstract subjects, we find even fewer compound words. But when we turn to the sports page and other kinds of informal writing, we Lire likely to find more. Indeed, since ME, the numher of apparent compound types has seemingly increased, if we merely consider the composition of examples from various parts of speech. These types, for example, arc cither infrequently or never found in O t poetry and prose: N N V V V Adj.

+ + + + + +

V V-ing N Prep. Prep. Adj.

= = = = = -

V babysit Adj. man-eating N pickpocket1 V throw up N setback N deaf mute

Adj. Adj. Adj. Adj. Adj. Adj.

+ V + Prep. + Prep. + V-ing + N + N

= V roughgrind = V blackout = N blackout = Adj. easygoing = V stronger m = N redcoat1

1 The compound type pickpocket does not follow the OE pattern where the second noun names ihe class of objects referred to. Thai is, л harstfty is a kind of fly, but a pickpocket \i not a kind of pocket, hut rather a kind of person. ]t did noi <.vxur in English until the I4ih century, probably under the influence «Г Frenth. 2 The type redcoat is like pickpocket In thai a redcoat is DOt a kind of coaf when it refers to a person. Compare hunchback, pah/ace, redbreast. These are rare before the ] 6th century. (91, Ш , 136J Moreover, something much like поил + noun compounding that results in a structure sharing characteristics of both a compound word and a phrase is occurring more frequently than at any time since OE, particularly in American English. Many writers of moderately technical or bureaucratic prose now quite unselfconsciously pile up strings of nouns instead of putting them into fully articulated syntactic structure*. For example, a system that alerts astronauts to emergencies involving the impact of meteorites travelling at high speed becomes a high speed meteorite impact emergency astronaut alert system. Й.З: (l) What would be the term for the board on which is posted the schedule of maintenance for the room in which planners confer about this system? The hook on which the board hangs? The salesman who sells the hook'.' {2) Iь there any general principle by which these strings are constructed ? That is, given the fully syntactic construction: The salesman of the hook for the board on which someone posts the schedule of maintenance for the room in which conferences are held byplannersof'thesystem that alerts astronauts to emergencies involving the impact of meteorites at speeds that are high, can we predict the

t24

WORDS AND MEANINQS

order of the noun-string? (3) How would we determine whether such a string is a new word or a syntactic phrase? Does the stress pattern suggest anything? PROBLEM 6,4: Compounds made out of noun 4- noun = noun constructions are common in all the Germanic languages. Examine these noun + noun = noun compounds. What sort «f problem do they suggest that goes beyond merely saying that noun + noun = noun? flower bed, wuierbed* steamboat, steartibatk, steamfnter, shrimpboat, sailboat, sailcloth, coalyard, coalsmoke, airgttn, shotgun, elephant gun, water pistol, horsepistol, horsefly, horseshoe, snowshoe, snowman, milkman, fireman, firehouse, fireplug, spprkplug.

Derivation Another strongly ruie-governed way we form new words in English is by adding affixes to the beginning or ends of words. Consider this sentence: Theexfensaus destructness of the atomive bomb exemplicates the unpassiblewtee task of defencement against it. It makes sense, but obviously something is wrong, -otts makes adjectives out of verbs, but ex tens- takes -ive, not -ous: extensive, -ness is a noun ending, but it occurs with adjective roots: boldness, happiness. Tn short, the pieces of words, the affixes, are not properly distributed. We inust distinguish here between two kinds of endings; inflectional and derivational arrives. For our purposes a derivational affix is any пол-independent element added to a word that changes its part of speech or its basic semantic meaning. For example, the plural s appended to boy does not change the meaning of the stein, boy. The third person singular -s does not change the meaning of walk. Nor does the possessive -s added to a noun, or the past tense -ed, the progressive -ing, or the past participle -ed added to a verb. But when we add -ize to final and get finalize, the adjective becomes a verb, -тем added at the end of the verb replace makes it the noun replacement. The -cy added to the end of captain does not change captain to another part of speech, but it docs change the meaning from the person to the abstraction. The -ish added to m/does not change it Lo another part of speech, but the meaning of reddish is different from the meaning of red. There are other tests. 6.5: Here are some words with derivational and inflectional affixes. Some are obviously incorrect. Use these data to suggest at least two more test* to distinguish them. 1. 2.

wtion-tiationality-nationaUt ies;nat ion-naUons-naUonsality just-justify-justifitdi just—justed—ju&iedifу

CHEATING NEW WORDS

3, 4, 5, <j.

125

reverse-reversat-retersals; reversv-reverses-resersesal sick-sh'keu-sirkening; sick-sickiug-sickingen y/alt-waiter-waUeT 's: wait-wait 'a- wait'ser (tmusetiQii-afnusence-amuseinent-, aniused-itmuses-amuswg entra tion-en trance-en tenrten (; entered-enters-en taring uction-actance-ectmettt; acted-acts-acting

Where do elements Listed as inflections occur relative to derivational suffixes? How ma ny inflections can be added to a word 7 How many suffixes ? Do other characteristics distinguish inflections from suffixes? Just as we have gained many new words through borrowing and compounding and lost many others through disuse or replacement, so have we changed how we add pieces to a word. Like compounding, affixation was a vigorous process in OE. But while a good many suffixes and prefixes have survived as fossils, not all of them are productive in MtidL in the sense that they can still be Freely combined with stems. For-, for example, was a very common prefix, occurring in hundreds of words. It intensified the meaning of a verb, as in forlorn (i.e., very lorn, or lost); от it could negate the verb, as in forswear. But beyond forbid, forlorn, fnrhear, forswear, forget, forgo, and forsake, we have no really active words in English that use that prefix. Nor can we add/or- to any new words: forkurt, fartake^/orotund, forwork. Jt is a relic form, a linguistic fossil. The OE prefix ge- that attached to the beginning of almost any verb has disappeared entirely, leaving only meaningless syllables in a few words like handiwork (from handgeweorc)* aware (from gewsr), enough (from genog), ajf'nrd (from gefortiian), and either (from zgfer, a contraction of a-ge-hwxifer). Ge- had a variety of meanings, among which was a perfective meaning with a sense of finished or done. If it were still active, we would say gebroken, to mean entirely broken up; or ge-eaten, to mean eaten up; oigeburnt, to mean burnt up. 1

PROBLEM 6.6: What, if anything, has ModE substituted for^e- ? But other affixes still thrive. Among the most vigorous of those inherited from OE, for example, is the -ed \afour-tegged, stouthearted, and so on. It is not the past tense -rJor perfect -ed. It comes from a Germanic suffix meaning "with." We can add it to -virtually any concrete object which belongs to something else; a steel-shafted club, a "Si-keyed typewriter, a slx-wheekd truck, a two-headed giant. 6.7: A list of most of the principal prefixes that have occurred in English since the OE period is shown in Table 6.2. AfKer the prefix is the source: F -= Modi:, F = French, T. = Latin, Gk = Greek. U there is no source listed, then the aflix is native OE. Following the source for a borrowed

126

WORDS AMD MEANINGS

affix is the approximate century in which it first occurred productively in F.nglish. A second date indicate* that the affix flourished much more strongly than at first. The examples are modern words if their prefixes are still found in ModE. If the prefi* is completely lost, an OE example plus its modern translation in parentheses follows: ed- m edwnden (reversal), for example, is completely lost. If ait OE prefix still occurs in a word but is unrecognizable, then a ModE word with whatever is left of that prefix, follows in square brackets; and-, for example, is unrecognizable in answer. As you examine Table 6.2t answer these questions: J.

2. 3.

4. 5. 6.

7.

8.

What, if any, are the semantic differences between OE prefixes and the borrowed prefixes? {Peif[ectice\) in the explanatory note means that the action is completed, as explained above. What proportion of the descendants of OE prefixes are still used today productively ? What is the general proportion of productive native English versus borrowed suffixes? That is, with occurs only in withstand, withhold, withdraw, withal/, within, and without. No other words begin with with. With- is entirely non-productive. De- meaning undo is extremely productive: de-Americanise, depressurize, defuse, and so о si. Are there any prefixes which have virtually become words themselves? Would it be possible to give up the productive native prefixes entirely? What native words can we substitute for each foreign import? For example, inter- could be replaced by between, so that international might be represented by betweenland. Would the answers to any of these questions be significantly different if we decided to call the following elements prepositions rather Lhan prefises? sfter, fore, ford, mid, ofer, on, Jo, frurh, under, up, ut, wiO, wider, Write a paragraph of ModE attaching OE prefixes to the appropriate ModE words. (97,136,175)

6.8: A list of some of the principa! suffixes that have been used since the early OE period is shown in Table 6,3. Some of the same information is provided after the suflsxes as in Table 6.2 on preli\es, plus whatever grammatical change results from the suffix. The native suffixes arc spelled as they were in OE, (1) What is the difference between the functions of suffixes and prefixes? (2) What OE suffixes are still genuinely productive? For example, -fast occurs in rotorfait, steadfast, shamefaced (originally -fust, but now incorrectly reinterpreted as -faced). But it occurs in very few oLher words, -ed, from the Germanic *-odi. meaning "provided with" (not the -ed past tense or the -ed past participle) is freely attached to almost any concrete noun; a round-windowed house, a well-penciled student, -ed is productive; -fas! is not.

CREATING NEW WORDS

TABLE 6.2

PRINCIPAL ENGLISH PREFIXES SlMCE OE PERIOD

Prefix (Source) (Meaning)

Date

1. a- (perf./intensifier) 2, a- (generalizes pronouns) 3. and- (against, toward) 4. ante- (L) 5. anti-{Gk.)

16c. 16 c.

6. d^-(Gk.butinOE) 7. auto- (L.)

19 c.

S. х- (without) 9. a?f- (perfective aspect)

10. ifter-

11. seg- (generalizes pronouns, adverbs) 12. be- (around/over; intensifier; N > V) 11 bi- (L.) 14. circunh (L.) 15, ™-(L.) 16. counter- (E.)

17.
19. ed- (again/back) 20. el- (from elsewhere)

21. e«/™-(F.) 23. ex/ra-(Lr) 24. for- (inteus./perf,, negiition) 25. fore- (precedence/ pre-emincncii) 26. ford- (motion towards) 27, fuU (complete) 28, gain- (against) 29. Se- (pcrf,/collective) 30. 31. 32. 33. 34. 35, 36.

hyper-(Gk.) гчн(- (intensifier) inter- (F.) infra- (L.) щя/^ (F.) 4w?a-(L.)

Ш. mid-

127

Example afysan (drive away) [await] ahwxdet (either of two) [either] attdsaca (adversary) [answer] ante-chamber anti-aircraft arch-thief autohypnosis zgikle (without payment) xfwyrdla (damage) afternoon, aftershock xgfiwn (everyone) bedeck, befriend

16c.

bi-vafoe 15/17c. circumnavigate co-worker 15 с 14/16 с. counterstatement

15/20 с. decontaminate disallow 14 c. edwenden (return) etland (foreign country) 14 c, encircle, embolden 19 e. ex-wife extra-terrestrial 15c. forswear foreshadow forthcoming fulfill gainsay gesceran (cut through) [enough]

15/17C. hyperactive inside infrod (very wise) international 14 c, intrapariy 19 с 15c. mafadapted 19c. metatheory 19 c. microbus mid-century

)28

"WORDS AND MEANINGS

TABLE 6.2 continued 39. mini- [L.) 40. mis41, mutti- (L.) 42. neo-(Gk.) 43. «on- (L.) 44. qjf- (perfective) 45. ofer46, o?;- (begin/reverse action on) 47. or-(without/Intens,) 48. off-(close to) 49. pan- (Gk.) 50. para-{Gk.) 51, /w/>-(Gk.) 52, post- (L.) 53. /JJ-О (F.) 54, pro55. /J™/O-(Glf.)

56. pseudo- (Gk.)

20 c.

17/19 с 14c.

17c. 19 c. 18 c. 17/19 С 14/19 с. 19c, 16 c. 14/16 с 13/15 с.

58, retro- (Gk.) 59. ram-(half)

14/19 с.

60, 61. 62. 61 64. 65. 66. 67. 6869. 70.

14 c.

semi-{h.) IBs- (extensive/lasting) sub- (L.) super-(L.) supra- (L.) fn- (motion towards) to- (perfective) trans- (L.) ^WWJ- (through/complete) ultra- (L.) unjinlimlil- (OE/F.)

71. under72. кя/-(Ь.) 73, y/j-

74. 75, 76. 77. 78.

i//eice-(L.) wan- (negative prefix) wiS- (a way/against) wiiJer- (opposing/counter)

79. j'fli£>(f)- (around/about)

14 c. 15c. 17/J 9 с 16/19 с. 19 c,

mini-skirt misbegotten mufti-colored neo-classical non-payment ofgyfan (give up) overpayment onbyrdan (incite), unbind orsawle (lifeless) odstandan (stand still) Pan-American para-military polydialectal postgraduate predetermine pro-communist proto-Germanic pseudfi-hsteliectual redassify retro-rocket snmworht (half-built) [sandblind, originally satnbHnd] Semi-circle siftnihte (perpetual night) subsurface super-heated supra-national tocyme (arrival) [toward] tobrecan (break up) trans-Atlantic throughway ultra-sonif unequal, insincere, impersonal, illegitimate understandan (perceive),

15/19 с. unicellular uphold outlay vice-chairman 15c. wanhai{$\t±) [wanhope] withhold, withstand wiSersaca (adversary) ymbgang (circuit)

CHEATING NEW WORDS

129

4

What is the percentage of retention ; What is the present percentage of native vs. borrowed productive suffixes? (3) Arc there any suffixes which have virtually become words themselves? (4) Would it be possible to give up the native suffixes entirely? (5) Play the role of Saxonizer again (an example of the productive use of -ize and -er). What native words, if any, could be substituted for each borrowed suflix'.' (6) What semantic information is conveyed by various suffixes? (97,137,175) TABLE': 6.3 Suffix

I. -able 2. 3. 4. 5. 6.

-асу -ade -age

7.

-art

•al

-al

PRINCIPAL SUFFIXES Source

Date

(F.)

14 c.

Example

V > A

breakable

confederacy Л > N 14 c. (F-) (Span.) 19c. N > N orangeadeN > N, V >• N baggage, passage (F.) 13c. (L) suicidal 14/18 с N > A V disposal 14c. > N (F) N N African 16c. > A/N > (M V > N (F.) entrance, existence 14c. N > N, A > • N presidency, hesitancy 14c. (L ) V > N assistant, president 14 c. N > N parliamentarian (L.) 16c. N > A fragmentary (L.) 15c.

8. -ancet-enee 9. -ancyi-ency 10. -ant]-ent 11. -arian 12, -ary 13. -ate (F-) 14. -athon (Gk.) 15. -bxre (productive of) 16. -bora (Eigetitive) 17. -burger (Gm.) 18, -cade 19, -cracy (M 20. -cund(has th(ъ•• nature of) 21. -cy С--) 22. -dom 23. -e 24. -ed (provided with) 25. •ее (F.) 26. •eer (F.) 27. -el 28. -eh 29. -en

Grammatical Change

13/16 с A > V, IS > V 20 c. V > M N > A

20 c.

N > N N > N N > N N > N N > A

activate, orchestrate talkathon litsibxre (agreeable)

mundbora (protector) shrimpbwger motorcade mobocracy godcund (divine)

14c.

N > N , A ?- N captaincy, accuracy N - N sportsdom, wisdom hlude (loudly) A > ADV legged N > A

15 c. 16 c.

V

=• N

V > N, N >• V > N V > N N > A

N

payee auctioneer, conventioneer dyret (hole) [swivel] byrgeis (tomb) woolen

130

WORDS AMD

MEA]S"IN<JS

TABLE 6.3—continued 30, -en 31. -end (agent) 32. -er 33. -ery 34. -ese 35. -esque 36. - m 37. -estre 38, -«/ 39. -cue 40. -/^sf 41. -/• 53. -tf 54, -iht 55. -isc 56. -ism 57. -jsr 58. -г7е 59. -ids 6 0 . -(/>•

61. 62. 63, 64. 65, 66.

-we -ize -tec -leas -Izcan -fe

67. -/
A > V V > N

13 c. 15 c. 17 c.

(F,)

16/19 e. N > N

14 с

(Ger)

20 c.

(Ger.)

20 c.

(L.)

> A, A > A > A > N > A > N N > N N > A,N > N N > N V > NT A > N N > N N >A N > N, A > N N > V N > A N > A N > A, A > A N > N , V > N, A >N К > N N > N, A > N X >N N > N, A > N N N N N

(F.) (F-) (L.) (L.)

18c. 17c. 15/20 с 15 c.

(Gk.)

14/16 с

(Gk.) (L.) (Gk.) (F,)

17c. 13/19 с 20 c. 14/16 с

(F./L.) (F.)

V > A 14 c. 13/19 с A > V,N > V V >N N > A Л > V 15 c. V > V N > N 16 c, A > A, N > A 15c, N >A N >N JN > N 20 c.

(F.) (F.)

68. -Дс

69. -like 70. /teg 71. -mania

V >N N > N, V > N N > A, N > N N > Л N > N N > N V > N

(F.) (It.) (H,) (Gk.)

(Gk.)

darken demend (judge) driver fishery, bribery Japanese KipHngesque goddess mobster.punster bsrnett (burning) [thicket] usherette hea(}fast, sieadfa.it fourfold songj'est, gabfest sinful jishfurter nationhood Austrian Bobby, daddy, Katie movie, quickie Shake spear iana alcoholic beautician, musician liquefy earthy porniht (thorny) girlish, reddish militarismt criticism, extremism behariorist Israelite, suburbanite put-it-off-itis Christianity, sentimentality assertive finalize, crystalize feajlac (robbery) friendless nealsecan (approach) sparkle piglet sicklyt kingly godlike princeling auiomania eetulnaed

CEEATIISfO NEW WORDS

131

TABLE 6,3—continued (F.)

14 c.

(L.)

20 с

-ory -off

(L,)

16 с

-он-*

(F.)

14 c.

(£p.)

19/20 с 20 с

(F.)

J4c.

72. -tttent 73. -nessf-nis 74. -of 75. -oriwn 76. 77. 78. 79. SO. 81. 82. S3 R4. 86. 87. S8. 89. 90. 91.

-s -sian -. -l -ung -ure -v -wende -v

20 c.

V A V V

> > > >

N N A N, N > N

V A N N A N N P A N A V V N A N A

> > > > > > > > > > > > > > > > >

A N,V > N A N ADV N V P Л N A, N > A N N ADV N ADV, ADV

amusement weariness pattcol (thoughtful) corsetorium, printutorium obligatory length, growth poisonous hatred thereabouts, days friendship mxrxiun (proclaim) flflWflfjr lonesome laimdreteria oldtype, Chicago-type ruttttaig pressure homeward Acr^fHt/e (healthy) educaiionwise, happy wise

A kind of word formation like that we have just discussed but which is largely limited to the creation of scientific terms joins two elements, neither one of which need be a complete word. These are called hound forms because generally they must be attached to another word or to another btjund form to make a whole word. F o r example, we can take almost any of the elements in the first list below and attach it to almost any element in the second and get a very impressive result: 1.

2,

electro-, duple- (doubling), hydro-, eikono- (image), hetero- (different), cephalo- (head), atidro- (human), idSo- (individual), proto- (original), synchro- (simultaneous), phyb- (racial stock), toxt- (poison), pyro- (fire), vivi- (alive), -dasia (fracture), -glyph (carving), -crania (of the head), -odym (pain), -chronic (of time), -graphy (writing), -genesis (creation oQ. -'<™e ( t o n e ) > •metric (measure), -neural (of the nerves), -ptex (fold), -tropic (turn towards), -vora (to cat), -plasm (molded). (161)

Some of ttese prefixes could have been listed in the previous lists. In proiagenesis, for example, the genesis is. now also a word, makingproto- a kind of prefix. The same is true with hydro- as in hydroelectric and so on. So in

132

WORDS AND MEANINGS

many cases, there is no real distinction between many of these and those Jisted in Table G.2. PROBLEM 6.9: Translate ten possible combinations into native English coinpounds. Kor example, pyrogenesis can be translated mtofirebirtk Then test them on someone to determine whether they can be understood in the sense of the original Greek or Latin word. Is there anything thatpyrogenests communicates thatfirebirth does not? There was a time when one of the serious questions among critics of usage was whether it was proper, correct, decent, to combine stems from one language with affixes from another. Scientist was once condemned as an unspeakable hybrid, composed of a Latin stem: sclent-, plus a Greek suffix, -ist. One is tempted to wax Freudian on the racial implications ofthe attitude reflected in the term "hybrid" and why some writers on linguistic propriety were overcome with revulsion at merging elements from different languages. Appropriately enough, the word racial itself was a target: it combines a Latin suffix with an Italian stem. It is an attitude that recalls some aspects of the sixteenth-century Saxonists, In answer to such critics, we can only cite words so completely naturalized that objecting to them on principle would be simply foolish. Beautiful is a French stem with a native suffix, tidal a native stem with a Latin suffix. All these are of mixed parentage; peaceful, successful, resentful, fruitful, merciful, doubtful; gentleness, faithfulness, promptness; paternally, royally, modestly,fortunately, certainly, allegedly; breathable, understandable, unthinkable, unbreakable; merriment .shipment; winterize. Like most questions of this kind, the strictures of those who would improve by purging are ignored by those who, as they use the language to express their thoughts, shape it to meet their needs. Jf a "hybrid" results from someone's expressing himself as naturally as he can. then the fact that he creates a hybrid is irrelevant. Some mixed creations, of course, do grate against the ear: unusualness, drinkability, responseful. But we might wonder whether these grate because they are hybrids, or because they arc unfamiliar, or because they violate the rules of combination. Redetoxificationability is a Latin blueblood that follows all the rules, but it is grotesque. Whether such words bccfime part of the language really depends little on prescriptivists and less upon historians, but almost entirety on how many speakers decide they want to use the word. 6.10: (1) What requires a greater intimacy of linguistic penetration of one language into another, attaching a native suffix to a borrowed word: graceful, spiritually, correctness, or attaching a borrowed suffix to a native -word: beatable, amazement, oddity ? (2) How would dating the patterns PROBLEM

CREATING NEW WORDS

133

mixed constructions support or weaken your guess? (3) Which are less restricted as to the specific words they can occur with, prefixes or suffix»7 For example, how freely can we combine re- meaning again (аз in reset) with verbs V How freely can we combine -men! with verbs (as in amazement)'! How does this apply to the discrepancy between Lhe number of borrowed and native prefixes we have in the language? affixes ? PHOBLEM 6.L1: Review the compounding chart (Table 6.1), Must we revise our description of compounding in order to take derivation into consideration? That is, what happens if we treat -ing (as in oil painting or easy-going) and the perfect -и/(or its equivalent, as in rough-ground or brokenhearted) as derivational endings? In what ways would we have to refine [he history of compounding? Creating by Analogy1 Backformttiiims When a noun like lamentation or obstruction or revolution was borrowed into the language rather than the verb forms foment, remit, or obstruct, it was virtually certain that the language wouid eventually increase not by one word but by two. For if we borrow transmit and know by the rules of affixation that we can create transmission, then a. native speaker who comes across exclamation knows, without having seen exclaim, that exclaim is, by the rules of the language, a possible Fnglish word. Given amiable, he might not know whether the noun is amiability or, following his native instincts, amiableness. But given either of those forms, he knows that amiable is possibie. Oncc a principle of proportion is established (x - a: x as у - a: y), speakers easily extend it even to words such as these, which, strictly speaking, do not qualify: pedlar, beggar, hawker, stoker, scavenger, swindler, editor, burglar, sculptor. For these words, the proportion was incorrect because none of them had an active verbal root preceding the -erjar ending. Pedlar and hawker are of unknown origin; beggar is from the name of the mendicant order, St. Beghard; burglar is perhaps from the OE root burg-brer tian, or house/room-breaking. Scavenger is from the French for toll collected in a city and is not a verbal root. Editor, stoker, and sculps or have historical verbal roots, but they were not active when English borrowed the nouns. These are usually called hackform;itiorts.

PROBLEM 6.\2: These words were also borrowed directly into English with, originally, no verbal derivation associated with them. What modem verbs have we derived from thent? reminiscence, resurrection, pre-emption, ririsectian, electrocution, television, emotion, donation, enthusiasm, jelly, peevish, aggression. What about these words: motion, cognisance, contempt, ambition, incursbn, infraction, abeyance, imprecation, transition, insurrection, susceptible, insuperable, inquisition"! Should we have verbs for them ? What would they be ?

134

WORDS AND MEAMIISfOS

Related to the agentive back formations (beggar-beg) is a class of words we have already looked at, buL which historically is rather new. Compounds like firefighter, babysitt&, housebreaker, bricklayer, and so on, compounds with а поил + тегЪ-ег pattern ate very oid, dating hack at least to the ninth century. But rather more recent is the verb compound related to them without the -er ending. A few of these verbs date back to [he thirteenth century: backbite, for example, which, at least according to the Oxford English Dictionary citation dates, followed backbiting (first citation 1175) by perhaps 125 years. But it has been only in the last few centuries that the form has become genuinely productive. Many, if not all, of these new verb compounds seem to derive from baekformations of three types: (1) N + V-er: hootlkker-to boo! Иск, (2) N + V-ing: joyriding-to joy ride, (3) N + V-ed: henpecked-to henpeck. M 6ЛЗ: Here are some examples. Which of the three types of verb compound seems to be the most likely source for each? Check your tonelusions against the dates ciLcd in the OED and its Supplement for a sampling of these, housebreak, shoplift, sightsee, housekeep, stagemanage, moonlight, typewrite, typeset, backpeddie, hog tie, drydesn, bootleg, sleepwalk, hitchhike, ghostwrite, pinchMt, proofread, sharecrop, babysit, chainsmoke, copyread, playact, handpick, doubtepurk, brainwash, breastfeed,forcefeed. 6.14; Create ten new compounds of the noun-object + тегЬ type directly: to hoitserepuir, tobirdfeed, to haircut, and so on. How natural do they seem? Would this be an argument against someone who claimed that words like bookbind. hand-tew, housepaint arc not back format ions but rather direct compound creations? How would this influence on r description of compounding in Engl is h?

Tn some earlier problems, we have seen how rich English is in ways to convert one part of speech to another by adding suffixes: decide-dedsiou, sing-singer, orchestra-orchestrate. But one very common way to convert one part of speech to another is for a speaker just to use a form that represents One part of speech in the position of another without changing the form of the word at all. in effect, he adds a zero-suffix: 0, For example, someone who knew he could say The tights gleam in the night, used gleam ан a noun, as in a sentence like lean see the gleam in the night. A similar process created verbs out of adjectives: / made the desk clean—/ cleaned the desk, and nouns out of verbs: / looked out of the window I took a took out of (he window. All of this speakers can accomplish without adding a suffix. While this has always been a practice in English, Shakespeare and other Elizabethans were particularly notable for ihcir free use of such conversions:

CtfiATlNG NEW WORDS

135

I warrant Him, Petruckio is Kated (Tarn inft of the Shrew, Г1Г i i.), He Devill-Porter it no further (Macbeth, IT.jji.), "Pis . . . suck stuffe as Madmen / Tongue and braine not (Cymbeline, V.iv.), He words me Cyrles, he words me (Antony and

Cleopatra, V.ii.). This is such a widespread phenomenon that most common content words of English occur as more than one parL of speech: man, hand, foot, eye, doud, mouth, light. But this may be a misleading way of speaking. To be consistent, we may have to distinguish between a linguistic form1 and a word. Walk is a linguistic form. It can occur as the word we use as a verb: He walks, or it can occur as the word we use as a noun: He took a walk. When we talk about the word walk as cither a noun or a verb, we fail to make a distinction that we have to make in other contexts. Is the word horn (as in an orchestra) the "same" word as the word horn meaning bony material? We can resolve the confusion by distinguishing between linguistic forms and words. If English speakers have for centuries converted one part of speech to another, they seem not to have felt the same about the suitability of all conversions. A good many conversions that existed earlier have disappeared in favor of another form—or of no corresponding form at all. PROBLEM 6.15: Here are two sets of words. (I) is a list of verbs formed from English and French nouns, (2) is a list of nouns formed from English and French verbs. Most of these conversions occurred between the thirteenth and sixteenth centuries. Is there any difference between the stability of nouns converted into verbs and verbs converted into nouns? Is there any difference in the stability of English and French conversions? The proportions are roughly accurate, 1.

Nouns Into verbs (English): /wrt, worship, child, gleam, rust, stream, master, hook, stone; word, den, blast, spark, snow, witness,fist, edge, length, head, spire, sleet, fellow, guest, chill, hammer, pin, wire, church, ring, moan, churn, filth, plough, shovel, weather, doom, knot, tail, arm, shuttle, roost, sponge, burden, riddle, .фаге, crank, gossip, chalk, soap, whore (50 words). Nouns into Terhs (French): sacrifice, curtain, fine, penalty, issue, print, treason, cement, tone, treasure, virtue, retail, exercise, plumb, dart, buckle, lure, beauty, license, robe, difference, plea, nurture, marshal, .flavor, rioet, solder, liberty, farm, acquittance, accrue, experiment, motion, brush, letter, trick, safeguard(37 words).

2.

Verbs into nouns (English): chew, keep, grind, steal, weep, dread, have, miss, stink, suck, look, feel, choose, hide, secrete, put, shove, crack, mourn, understand, warn, wish, blink, chide, pull, wink, build, write, skulk, fan, hunt, shake, stir, walk, stare, lap, bruise, freeze, rise, shoot, wash, like, befall, creep^grope, run, talk, clip, shut, bite (50 words).

I The term lexeme has also been used.

t36

WORDS AMD MEANINGS

Verbs into nouns (French): summon, escape, avow, appease, crush, disguise, pray, fescue, aim, tax, agree, arrive, elect, enforce, increase, rehearse ^ rush, support, depose, launch, piss, defence, enter, mock, praise, esteem, restrain, strain, acquit, approach, avoid, complain, discharge, furnish, presume, rejoice, remain, maintain,flourish, redeem (4G words), (J 25) PROBLEM 6.16: Here are some early French/Latin borrowings. Why would a Speaker of ME be unlikely (o convert these directly with no formal change in spelling or pronunciation into verbs: arrival, guidance, improvement, organization, reference, referenced If you had to speculate, wtmld you say that direct conversions arc more likely among native or arming imported words? Why? Words (or forms) have been changing their part of speech since OE, but a pattern of change which has suddenly flourished in recent times after its fourteenth-cent игу origins is the conversion of the verb + particle combinations like hold up (to rob), set back (delay) to the noun class with the typical compound word stress pattern: holdup, setback, layup, rundown, write in, and so on. PROBLEM 6.17: Review the compounding chart (Table 6.1) again. How would we have to refine our description of compounding to take into consideration both derivation and conversion ? Words from Names A very special kind of conversion occasionally combined with derivational suffixes involves only a relative handful of words in the language, but they have some of the most interesting histories of any of our words. These are words made out of proper nouns, words from names for people and places. Strictly speaking, many of these are not conversions since a word like watt, derived from the плпк James Wait, is still a noun when used to mean the measure of electricity. But since many others do involve grammatical change, we will discuss them all here. In some cases, we can instantly recognize the source: machiavellian, spartan, odyssey, quixotic, shyiock, jutias, pla tonic, sodomy, hamburger, frankfurter. In other cases, the original referent has disappeared from our lives or the word lias changed so much from the original that all sense of its source has disappeared: guy, gin, canter, pants, copper, money, slave, bedlam, place, weiner. These groups of words usually have a relatively short half-life compared to most derivations and «inversions. These nest words, for example, have all been used in earlier histories of the language to illustrate how we derive common from proper nouns: piimsotl, belcher, victoria, lewisite, hansom, pinchbeck, bant, lisle, daguerreotype, tram, andersvn, morrison, burke, brum-

CRUSTING NtW WORDS

13?

rnagem. Almost all of these Eire at best only dimly familiar as something out of the past or—more likely, perhaps—entirely unknown. Words arc derived from proper names for about the same reasons they are borrowed. Recall that we have distinguished between borrowing and using words like ketchup (the brine of pickled fiih, incidentally) and words like taboo. The word ketchup is necessary if the stuff is on the table. The word would disappear if the product became unavailable, with no consequent loss in our ability to express our feelings and ideas. This is exactly what happened to words like bant, tram, and lisle. For most of us, what they refer to no longer exist*. Having no need for the words, we have allowed them to fall out of use, just as wergild and scores disappeared. Taboo on the other hand refers to a significant area of semantic space in our culture, a space that had no exact name, perhaps, until we borrowed taboo. It refers not to an object, but to a way of behaving, a way of thinking and feeling and responding that is deeply embedded in our most basic behavior and attitudes. Thus it is difficult to imagine any circumstances in which a word like taboo would become useless and disappear from our vocabulary. Here are some of the words from proper names that refer to objects or elements that are so specific and so limited that they have not become "psycho-cultural" words: watt, ampere, farad, ohmt volt, curie, bvnsen, fahrenheit, derrick, silhouette, zeppelin, shrapnel, derringer, pistol, davenport, bowler (hat), derby, stetson, cardigan, mackintosh, pants, condom, bloomers, ietis, bikini, calko, cashmere, millinerу, damask, china, cologne, copper, sandwich, tveiner, frankfurter, hamburger, cereal, gin, bourbon, sherry, sauferne, champaign,xerox. PRORUGM 6.18: Can you generalize about the kind of referent most likely to acquire its name from a proper noun? Look up these words. What kinds of proper nouns are involved ? PRQULEM 6,19: These next words have become part of our "culture" vocabulary. They all originally referred to one specific place, individual, or group of individuals, real, fictional, or mythological- As you read this list, determine when you can the nature of the original referent that made it a source for ihe more general concept. You will have to refer to a dictionary for many of these words. Nouns: dunce, guy, maverick, lunatic, genius, hero, slave, assassin, thug, nemesis, quisling, mentor, solon, philistine, amazon, ttirk, chauvanism, masochism, sadism, enthusiasm, sodomy, onauism, cupidity, comslockery, fary, grace, bahel, blarney, bunk, panic, nightmare, sacrifice, odyssey, bedlam, harmony, music, mystery, magic, fate.fortune, mint, money.

t.i/t

WORDS AMD MEANINGS

Verbs; mesmerize, tantalize, vandalize, bowdlerize, gerrymander, canter, meander,pander,shanghai, boycott, hector, lynch. Adjectives: stentorian, spartan, Utopian, herculean, gargantuan, lethal, machiavellian, rabelasian, iiliiputian, quixotic, erotic, titanic; platotiic, tawdry, maudlin,jovial, (J60)

Gra лила Meal Wnrd I cuniiifion: A Summary At this point it should be c b r that arbitrarily dividing grammatical processes of word formation into mutually independent categories of compounding and derivation (including conversion and backformation) oversimplifies the problem. Both processes are at work in creating many words like easy going, ironing board, rough ground, nouns like output and setup, and verbs like hoidog (as in skiing), spotlight, and babysit. Grammatically-based word formation must be understood more as two interlocking processes or sets of rules which together or individually may generate words. The history of the process is then pbrai^d in terms of how the rules for compounding and derivation have together changed through history. Although a detailed analysis of those changes would be out of place here (indeed, it scarcely exists even in outline form yet), we have suggested the outstanding changes. The flow of the processes might be represented like this:

Grammatically generated vocabulary^

We begin with a semantic structure to be com тип ica ted. To represent that semantic structure, speakers can occasionally choose which path to take first. FireMrtti and pyrogenesis, for example, are not synonyms, but at one level of meaning, their semantic structure к very similar. The derivational route overlays the meaning "scientific" on pyrogenesis while the compounding route overlays the more general meaning of "poetic" or "expressive>L onfirebiri/i. In other cases, a speaker must follow one or the other path: Only doghouse, for example, will serve to represent the semantic structure behind that term. No unselTconscious derivation lets us produce something on the order of canicik (though caniculture, caninal, and caninity, have been used for dog-raisiitg, doggy, and doggisltness).

CEEATFXtj NEW WORDS

139

The fact that arrows point in both directions between compounding and derivation indicates that word formation often allows recursive functions. That is. we can compound: outhousei then derive: superouthause", then compound again: superouthouse salesman; then derive again: siiperouthousessSesmanship-, then compound again: jitperouthouse salesmanship award; then derive again: sitpetouthottsesalesmansliip nwarder, and so on. Or we can first derive: seller; then compound: booksellers and so on. Rule changes in the past 1000 years allow a few new compound forms: pickpocket, throw tip, deafntute, roitghgrind, redcoat. One permissible sequence has been lost: Adjective •+ Noun -^ Adjective (glied-mocl-g] ad hearted). A more general change is the loss of many native prefixes that communicated semantic information, particularly in regard to the state of an action indicated by a verb. That function has been taken over by a preposition like up or the auxiliary verb Ласе, Many borrowed prefixes have been introduced that very regularly correspond with native prepositions: interperiod = between times. Many borrowed suffixes came in with borrowed nouns, verbs, and adjectives, and for the most part, they continue to co-occur primarily with borrowed words. Л very significant change is the marked increase in our tendency to convert one part of speech to another without any affixation: A person who chairs a meeting can suggest that a motion be tabled and floor his opponenc. A unique kind of rule sequence was introduced with the simultaneous соmpound-derivation formation of forms like backbiting typesetter, or hctndpicked followed by the backformation of the verbs backbite, typesec, and handpick. We have also introduced in a somewhat jocular sense units that share the characteristics of both root words and derivational suffixes: -burger, -flitter, -athoti, -mania\ -omtfrt, -ferta, -type.

KOOT MOniFICATION The second large category of word-formation is illustrated when a speaker consciously or unconsciously manipulates the phonetic shape of words already in the language. Frequently, these new words will exist side-by-sidc with Lheir sources, but with a slightly different denotative or connotative meaning.

Clipping and S

The most frequent kind of change creates new words by clipping a part of a word, leaving only a piece of the old word. (The technical term for dropping an initial syllable is aphesis {defend > fend); for a syllable inside a word,

140

WORDS AND HEAN"ING£

syncope (segeder > either); fora final syllable, apocope {master > mas).) But frequently, it is a piece that goes on to become a new word in its own right. The earliest clips go back at least to the sixteenth century, probably much earlier: coz from cousin, gen! from gentleman, nius from master, and chap from a word we have now lost except as a name: chapman (salesman). 6,20: Here are the original forms for some clips. What words have we derived from them? Keep in mind that the resulting meanings will be fairly close: apply, brandy wine, defend, hackney, example, disport, periwig, cabriolet, omnibus, automobile, caravan, adrertisement, amend, doctor, attend, pantaloons, defence, veteran, laboratory, emy, mathematics, gymnasium, examination, mistress, dormitory, professor, estate, professional, fanatic, wizard, submarine, distillery, photograph, geneva, escheat, customer^ influenza, espy, benzedrine, acute, Chevrolet, etiquette,fadaise, despite, cocaine, co-education, affray, history, espiee, university (two clips from this one), alcoholic, quadrangle, accloy, graduate, estrange, penitentiary, rumbullion, abet, nuclear, turnpike, psychology. a(d)renture, sergeant, helicopter, convict, umpire, referee, communist, bookmaker, confidence man, women's liberation, detective^ homosexual. (I) Generalize about what part of a word LS retained and why. (2) In some cases, the clip and the source still nifer to the same thing: dormitory-dorm. Is it correct to say that the two words are identical in meaning? Is there any difference between them? Are dorm and dormitory one word or two? PROBLEM

Shortening, of course, also operates on compound words and longer syntactic constructions, but the outcome is in some ways quite different. Mob comes from mobile vulgis,pub from public drinking house, zoo from zoological garden, movies from moving pictures, narc from narcotics agent, (Coinci dentally, there is a much older wnrd, nark, for a police informant in general.) More often, the shortening of the phrase leaves an entire word, often the adjective modifying the deleted noun: Through repeated association, for example, private comes to stand for prkate soldier. The single word replaces the meaning of the entire phrase. And when this occurs, of course, the adjective becomes a noun, an automatic conversion, PROBLEM

6.21: What are some other clip-conversions of this kind?

Blends Another source of word formation which combines clips and something not entirely unlike compounding is called blend ing: smog, brunch, twirl, smaze, chortle, motel, sprig, gerrymander. Each of these is made up of the first part of one word and the second part of another: smoke + fog = smog, breakfast

CREATING NEW WOUDS

141

+ lunch = brunch, twist + whirl = twirl, smoke + haze ^ smaze, chuckle 4snort = chortle {in this one, the second word has been inserted inside the first), motor + hotel = motel, spray + twig = sprig. This sort of word creation is common in the prose genre that includes Time maga2inc and various nationally syndicated gossip/society columnists: mfmiidpating, blobstetruian (for a reducing adviser), stupidenfs (for student demonstrators), representative (someone to "rap" with), blaxploitation (exploiting blacks). Blending has supplied relatively few words. (165)

Acronyms Another form of word creation resembles blending, but is considerably more useful in a bureaucratized, politicized, advertised society where organizations and movements have names too long to fit conveniently into a commercial or a headline. When talking about the Congress of Racial Equality, the National Broadcasting Company, the North Atlantic Treaty Organization, the American Federation of Labor- -Congress of Industrial Organizations, we find it too difficult to give the full name every time we mention it. So we shorten them to their letters: Some are pronounced as letters: AFL-CJO, NBC, KKKt LSD. Others are pronounced as words: CORE, NATO, HOD, and so on. Most acronyms refer to organizations, but a few refer to scientific or technological referents: radar = radio direction finding and range: laser = tight amplification by stimulated emission of radiation; lox = liquid oxygen. Very few of these have entered our '*cultural" vocabulary. Perhaps the only ones to transcend the specific original referents arc from World War W.gestapo = Geheim Stoats Polices; nazi = Natiorialsoi.ialistiche Deutsche Arbeiterpartei; flack - Ftieger Abwehr Kanon (anti-aircraft); and perhaps snafu = situation normal, all fouled up, though this last one is now mildly archaic. Historically, the principle of acronyms goes back at least to medieval times when words were invented as mnemonic devices, much as a student today might memorize reygbis to remember the order of colors in the rainbow. A similar principle operated in the Middle Ages when Jewish names were blended out of the initials of a title, a personal name, and a father's name: Rabbi Schclomo Jizchaki becomes Raschi. In the nineteenth century, chemical substances were named by blending the initial syllables of their constituents: aldehyde, for example, from alcoholdehydrogenatum. (136) PROBLEM 6.22: What conclusions about conversion do you draw from these sentences? Can i hate your OK? J already o/e'd it. Then It goes OK now ? Yealt it's ready ок. OK, let's get going, ox! (OK. incidentally, docs not come from an 1840 political slogan, The People is OH Korreet; or from a Choetaw word,

142

okeht meaning it is so; or from the name of a Sac Indian chief. Old Кеикик; от from the initials of an Irish freight agent, Obadiah Kelly, or from Aux Cayes, a Haitian port; or from any numher of other sources. It seems to be from Old Kinderlwok. Kinderhook was the name of Marl in Van Burcn's (1782-1862, eighth president of the United Stales) native village, a name which was made into an appelation for Van Buren, then into a political rallying cry, then into the name of a political organization: The O.K, Chib. When the members of this organization got into a brawl with an opposition club, OK. became a kind of password,) (145) 6.23: NASA (the National Aeronautics and Space Admsnisiratiott), pronounced as a word means nothing more than the meaning of the words it stands for. How do you suppose the genesis of these acronyms differed from

that of NASA 1 HOPE, HELP JOIN, WA VES, NO W, PUSH1

ROOT CREATION

Direct Sound Symbolism Our last category of word formation is in some ways the most basic kind, quite unlike anything we have discussed so far. Harlier, we said that one characteristic of human languages which distinguishes them from many animal languages is that the acoustic image of a word does not resemble the message it conveys. That is, a bee encodes how distant a flower is from the hive by dancing along a scale of intensity that inversely matches the scale of distance from the hive. The more intense the cry of danger from an ape, the more intense the danger. Except when we raise our voices afong a scale that matches a scale of anger, surprise, fear, and so on, nothing in our language seems to have any unequivocal iconic correlation lo conditions in the objective world. We do not, for example, have words like ЫооЬ, which, when uttered slowly ibioob, means slow, and when uttered quickly: bioob^ means fast, or alternatively, high and low, weak and strong, fat and thin, and so on. A fundamental characteristic of human speech is that it is arbitrarily symbolic. We do not call a striped horse zebra because it looks like the sound of zebra. Now, after making that generalization, we must immediately qualify it, because there may he some limited areas of our vocabulary where sounds in words may correspond with the sound of their referents in more than an accidental way. The most directly onomalopoelic words in the language arc those which attempt to represent ejaculations. Many have a fairly long

CREATINQ NEW WOKDS

143

history: ouch (first cited in 1654), hah (1000), hofio (1150), phon(ey) (1673), ugh (1765), hmnmtnm (1854), pAew (1604), ;Н,Л (160S)b Wrw/; (1686), гдА (1894). Soine not eked in the OED are unhunh, yippee, tsk Mfc. Ttehee goes at least as far back as Chaucer's Alisoun in "The Miller's Talc": "Tehee!" quod she, and clapte- the wyndow to. And Absolon goeth forth a sory pas. Fragment A, ] I; 3741-2 Tush and wish we can find in the fifteenth century. More recently, wow was the all-purpose ejaculation. PROBLEM 6.24: The representation of naturalistic conversation at a very colloquial level is relatively rare outside of a few early plays from the fourteenth and fifteenth centuries. What does this fact have to do with the dating of these ejacu lat ions ? PROBLEM 6.25: What do you make of the way these interjections are now being used: (I) You can't pooh-pooh ihat, (2) You wowed them, baby. {3) We hurrahed him home. (4) I hate a rah-rah attitude. (5) I don'I like a lot afho-homg around here. (6) I got anowie, mommy. One step away from representing in words the sounds we make with our vocal apparatus is representing in words non-human sounds. PROBLEM 6,26: The Oxford English Dictionary lists these next words as "echoic" or "imitative." Comment, clatterJtiss,pop, siz;le, buzz, hum, bump, squeak, crash, snort, sab, howl, whistle, bieat, snicker, snore, mart, roar, boom, twitter, jabbert flash, fuss, throb, dump, jerk, job, knock, pat, splurge\ squelch, blab, blurt, jump, Mi,fitek,film$y,JUp, gag, gush, gulp. PROBLEM 6.27: Write down new words for the following sounds: (1) chalk screeching on a blackboard, (2) a suitcase bumping down a flight of stairs, (3) tinkling crystal, (4) someone sniffling, (5) a door slamming, (6) cloth ripping, (7) trees rustling in the wind, (Я) automobiles crashing, (9) an electric motor humming, (10} the horn on a diesel locomotive, t i l ) someone typewriting. Are there any generalizations to be made? YOJ will have to combine your results with your fellow students or yourself take a large sample to get enough data. PtOBLEM 6.28; What are the differences in the sounds referred to by bongbank, thrum-thrump, clang—clank, clurtk-thtmk, smack-thwack. Duiit, dindle, and (Л>/аге British dialect words which refer to sounds. Before you look them up in the OED, guess what they refer to.

/44

WORDS AMD MILA.NIMOS

Svncsthedc Sound SyitihttUsm

When we move from words referring to sounds to words referring to olher dimensions of physical phenomena, we are on even less certain grounds. The iconic dimension is still relevant, but now the relationship is synesthetic: Lhe use of one sense, sound, to represent another perceptual dimension—sire, movement, hue, emotion, for example. In "echoic" words, sounds arc used to represent sounds: but many psycholinguists have tried to made a case for sounds representing size, movement, color, and even emotions. All physical experience can be scaled: movement, dimension, sound, color, tactile consistency, temperature, and so on. Most have several simultaneous dimensions: sounds are measured by pitch, volume, overtones; color by saturation, intensity, and hue; dimension hy height, breadth, and depth; movement by velocity, acceleration, and direction. The problem is to dc[ermine whether any of these scales can be consistently correlated with any of the scales we can devise for the way we pronounce words referring to those phenomena. PROBLEM 6.29: Figure 6.1 shows pictures of four animals. Ask several people LO make up names for them without using parts of recognizable words. That is, do not call (a) something like a mmaapkant. Collect the names and tubulate the length of each word aeeording to the size of the animal, and the kind of stressed vowel. That is, if (a) were called a btorgey, the stressed vowel would be an *h aw ' or " о " sound. Use Table 6.4 to tabulate the sounds. High means a sound made with the tongue bunched up high in the mouth. Front means the tongue is pushed forward. Thus a high fronL sound is Щ or /i/ as in Jit and/etf/. A low back sound is /a/, as in bawd от bought. TABLE 6.4

VOWEL CHART Front

HiGH

/F/ feet

MID

№

LOW

fit

bait /e/ bet /*/ bat

Central

Back

M

la} pool /W pull /0/ boat

/a/ hot

N

bought

These are diphthongs, sounds made of two parts: /ai/ as in hide, /au/ as in !>4hi. js'tj as in bay.

PROBLEM 6,30: Here are descriptions for several actions. Ask several native speakers to make up words for them. Can any generalizations be made in

(С)

(OJ

Figure 6.1

146

WORDS ANn MEANINGS

regard to the kind of consonant sounds they choose? The kind of syllable structure? (I) Rapid movements of the hands LO indicate nervousness, (2) slow and steady pulling a load uphill, (3) thousands of lights blinking on and off, (4) tickling someone on the back of the neck, (5) punching someone in the stomach, (6) a line rain, (7) an avalanch of snow, (8) fog drifting through the streets, (9) the circular movement of a mosquito in Lhe air. PROBLEM 6.3 f: Here are some pairs of words. One refers to a very small animal, the other to a very large one. Which is which? weeg-wog, gleepglaop, dobe-dabe, fain-fawn, spmt-spint,fozzl0-fuzzletseug (pronounced like the -oo- in tooky-scewg(as in ought). Use the same pairs to associate with very light colored and very dark colored animals; very stupid and very smart animals, very beautiful and very ugly ones, very slow and very smart ones. Generalize about the nature of the vowels associated with the various concepts. Why is the likelihood of the
CREATING NEW WORDS

147

Phcnesthemes Some linguists have gone one step further in phonetic symbolism. 6.35: Here are some common words beginning with various consonant clusters. Is there any sub-group among them that is related in meaning? If so, how would you explain it? PROBLEM

CL-: gl'ace, glacier, glad, glade, gladiator* gladiolus, glamour, glance, gland, glare, glass, glaucoma, glaze, gleam, glean, glee, glen, glib, glide, glimmer, glimpse, glim, glissade, glisten, glitter, gloaming, gloat, globe, gloom glory, glo$s>glotti3tglove,glow, glue, glum, glut, glut ton, glycerine, FL-: flail, flame, flap, flare, flash,flaunt,/fee,flee!', flick, flicker, fling, flit, float, flog, flop, flip,ffounce,fiourish,jiout.flow, fluent, flute, fluid. flume, flurry, flush, fluster, flutter, flux, fly, flat, flaccid, flatulent, flee, flint, floor, flour, flower, flunkey, SP-: spat, spatter, spew, spirit, spit, spout, spray, spume, spurt, sputter, spurn, SN-: snarl, sneer, sneeze» snicker, sniff, sniffle, snigger, sniffle, snoop, snore, snort', snout', snuffle, snuggle, snob, snub. •UMP, -UNK,-UG, etc.: bump, thank, hump, ghffnp, dump, clump, clunk, sunk, funk, hunk, gunk, frump, glum, crumple, junk, lump, limk, mump, punk,plunk, plump, rump, sump, stunt, stump, trunk, tug, ugh, mug, hugj dug, rug, tug, jugJug, pug, rumble, tumble, humble, fumble. Are any of these words echoic? Look up groups of these words in other languages. What do you conclude? Look up various initial consonant clusters in other languages. What do you conclude? Mueh of what we call phonetic symbolism in this area is still specific to the conventions of English. Linguists have a word for combinations like fl-, gl-, sp- and so on Lhat seem to have a kind of diffuse meaning: phonesthetnes. About all that can be concluded from them is that if English speakers coin new words, they will very likely follow the sound patterns that have become conventionalized for the referent. (16)

PSYCHOLOGICAL AND SOCIAL MOTIVES IN WORD CREATION We borrow, create, transform, and derive words in these ways for reasons that arc semantic, social, and psychological. Some words are created

14Я

WOKDS AND MEANINGS

and borrowed to fit new artifactual and culturally restricted phenomena: wombat, Uwisiti, WPA, mute!, smog. OLher words seem to cover an area of psychic space that, as it were, existed in our conscious or subconscious mind and thereby needed words to objectify it: taboo, gestapo, шиЫЫ, spartan. Words tike these did not originally refer to broad areas of our cultural or psychological lives. Buc oitce these words were made available and precipitated or crystallized those amorphous psychological areas into the consciousness of language, the words both reHccLed part of English/American culture and directed the attention of its speakers to the concepts they refer to, thereby making those concepts more accessible. The social consequences of borrowing, creating, compounding and so on, though, are equally important in the way we adapt Lo or create a social situation. If we postulate two poles: formal and intimate, we can roughly assign different kinds of word sources to different situations: More Intimate

More Formal

TS ati ve in heritance С lips and com pounds Underived forms

Fre nch, Lati n,
Compare two admittedly extreme examples: 1. 2.

/ think ft! cut''cross the quad back rn the dorm 'n kit the books far the psych final with my roomie. My decision is to traverse lite quadrangle to return to the dormitory to study for my psychology firm! examination with She individual with whom I share a room.

The general interaction between social class and style can be represented very roughly as follows: Higher Class Formal^^^ Casual

Lower CJass "^"Formal

•^cSaT1

^ ^

Intimate—•— —Intimate

That is, the frequency of features characterizing a formal style will in the (relatively) lower social dialeet class generally be shifted statistically down relative to the styles in the (relatively) higher class dialect, except perhaps at

CREATING NEW WORDS

149

the most intimate level where half-finished sentences, murmured words, and tone or" voice convey information in much the same way in hoth classes. But there occurs a frequent exception to this generalization. In some cases, the very formal style of someone in the most upwardly mobile group will display more forma! cues in Formal situations than does the upper middle class speaker in the same social situation. This gives us a relationship something like this: Upper-M iddle Clasi

Lower-Middle Class —Formal

Formal' Informal Casual" Intima Le

~^ •— . —•

Informal .Casual Intimate

This may explain why so many French (and later Latin) words shifted into more casual as well as formal English after social classes began to shift. The rule for the distribution of Romance words in various styles of uppermiddle class social dialects might originally have looked something like this: 1. 2. 3. 4.

Intimate: Native words^,! д _ я й-и Informal: Native words M t o A _ B Casual: Native word5^ в л _ в ^ Formal: Native words 3 S t j J

+ + + +

French words aet w French words^,, „,_*_г French words^^ w x French words a f r l i W-X-Y-Z

(This says nothing about the relative sizes of any of these sets. Л would certainly be the largest, however.) As a speaker moved from an intimate to a formal style, he probably successively replaced native English words with borrowings: Casual X > D, informal: У > С, formal: Z > B, or more accurately perhaps, increased the frequency of one category and decreased the frequency of another. But just as a child overgeneralizes and uses -ed too frequently—hugged. goedt singed—so perhaps the upwardly mobile Angio-Saxon speaker used too many French words too often where they did not belong, in less than very forniid contexts, by inappropriately using the largest subset of French words, W-Z, Lhat characterised a formal style of a particular social dialect. H may have been Lhis palLern of word choice that John of Salisbury had in mind around the middle of the twelfth century when he observed that it was Fashionable for English speakers to use French words in English conversation. In this way, French words worked their way into the informal and casual speech of naLive linglishmen. A complete study of word borrowing, derivation, and creation would

ISO

WORDS AND MEANINGS

probe much more deeply into grammar and phonology. Stress patterns are a lest for compound words. But how we assign stress finally depends on the grammatical processes involved in compounding, a problem much more complex than we have described here. Derivation includes more than just adding an ending: rerise-recision differ in lhe vowel change from pairs like resist-resistance. But we have discovered some important facts about our words and iheir sources. The wide range of borrowing and processes of word creation gives us a flexibility and exactness of expression rivaled by few other languages. Our lexical resources allow us to modulate our style, to express fine nuances of denotative and connotative meaning, in distinguish among social dialects, and even to control the rhythm of our sentences. Its possibilities range from the Anglo-Saxon monosyllabic simplicity of Lhe ki In the beginning was the Word, and the Word was with God, and the Word was God. The same was in the beginning wilh God, All things were made by him; and without him was not any thing made that was made. In him was life; and the life WAS the light of men. And the light shine th in darkness; and the darkness eomprehendeth it not. John 1:1-5 to the rolling polysyllabic Latinate chords of Dr. Johnson i That praises arc without reason lavished on lhe dead, and thai the honors due only in excellence are paid to antiquity, к a complaint likely to be always continued by those who, being able to add nothing to truth, hope for eminence froin the heresies of paradox; or those who, being forced by disappointment upon consolatory expedients, are willing to hope from paslerity what the present age refuses, and flatter themselves that the regard which is yet denied by envy will be at last bestowed by time. Preface to Shakespeare

OUR VOCABULARY AND ITS SOURCES: A SUMMARY

In Table 6.1. we sketched the outlines of the processes involved in grammatical word formation. Now that we have reviewed several other processes, we can suggest a more comprehensive system. There ure three sources of roots, three kinds of root-modifications, and an interlocking process of two kinds of grammatical word formation. The entire system can be approximated like this:

c u t ATI NO Nnw womns

IS I

RtJOTS Intucntcd J3ori\iwtfU

/

MODIFICATION

\_ Clipping

t 1

Ulnullnn Acronym)

t!ENE RATION

^Х

W

Deriretlon

) j

1 1

y-^

/ \ Mil- VOtAHL"LAKY OF ENGLISH Figure 6,2

PROBLEM 6.^6: Provide examples For the Following processes plus five more of your own device. The rool-source is in brackets; the process follows, 1. 2. 3. A. 5. 6. 7. 8. 9. J(l.

[N a live) [Created] [Borrowed] [Native] cli pped [Borrowed] clip ped [[Native] +• [borrowed]] blended [[Native] +• [borrowed)] compounded [[[Native] derived] + [native]] compounded ([Й arrowed ] cli p ] derived [[[Borrowed] derived] + [borrowed]dip]compounded.

Chapter 7 SEMANTIC CHANGE

Unlike changes in pronunciation, syntax, or usage, the ways in which meanings change open up such interesting bypaths on social, political, and cultural history that one is tempted to wander aimlessly through historical semantic space, turning from one etymological curiosity to the next. Surly, for example, conies from M К sir-/у, from Old French sire, which came from aeoir, from the slurred pronunciation of senior, the comparative of Latin senex, or aged one Л Sir tt.e,, a sire) apparently acted sir-ly. The re-spelling hides the fiict that English Sirs—or those who styled themselves such—in the fifteenth and sixteenth centuries apparently were not a generally likable lot. Christ, cream, cretin and grind are all related. An Indo-Еишреап root related to grind, *ghrei-, became in Greek khristna, or the anointing oil produced by grinding. Christ comes from khristas, past participle of Greek kftrein (to anoint), since he was the one anointed with oil produced by grinding. Chrisma (oil) also gives Old French cresnte, which when borrowed in ME became vretne, or the most desirable liquid that separates out from milk, or cream, any whitish semi-solid substance that is rubbed on something. CreUn is a Swiss patois word ultima (el >' from Latin Cltrisriunurn, which meant human creature, apparently a euphemism for those who suffered from the condition found so frequently in the Alps. Воре comes from Dutch doopen, to dip, hence any sauce that is dipped. 153

154

WORDS AND MEANINGS

Ft then came to mean any viscous fluid. Then in the Inter nineteenth century, it came to mean any opium preparation, then a specific one used on race horses, hence dope-sheei (information about horses), hence dope first as spurious information, then any information (i.e., the straight dope and to dope something out). The opium meaning then generalized to any narcotic, to pills, powders, or fluids, and also shifted to the person who is doped, a doped person, hence a dope. Airplane dope is a relic from the viscous fluid meaning. And in one of the more curious results of meaning change, it has been seriously proposed that whip, viper, and wife are all related. One could go on and on listing histories like these. But it would be only that—a list of anecdotes. It would give us no insight into the general processes of language, into any principles of semantic change, into how and why semantic changes occur, into the regular patterns of change and what we can discover through semantic change about our language, possibly abotit our society, and perhaps even about our mental processes.

THEORIES OF MEANING

As we have emphasized, however, we cannot explain how meaning has changed without an adequate theory of how words have meaning. Unfortunately, of all the areas of linguistics, meaning is the least well understood. Not only are linguists unclear about how a theory of meaning should explain what a word means; they are not even certain that when they ask what a word means, they know what they want to know. The meaning of "meaning*' has been debated at least since Plato, and the semanticists exploring that question in the last decade have raised more questions that they have answered. But as in physics, biology, and all the other lively sciences, progress in linguistics can be measured by how much we discover we do not know. Older theories of meaning have proved to be irrelevant to problems of linguistic meaning. Not too many decades ago, psychologists and philosophers talked about meaning as the images that words stimulated in their minds. Unfortunately, no one was able to describe these images in a way that let the listeners perceive how they might be similar to their own images. And under any circumstances, it is difficult to imagine what images, if any, spring before us when we hear a sentence like Relational concepts presuppose perceptual processes. When behavioral psychologists came to dominate psychology in the twenties and thirties (as to a great extent they still do), it became unfashionable to introspect about what went on inside one's head. They began quanti-

SEMANTIC CHAWE

155

tatively describing what a word meant in terms of what they could objectively observe about the way people behaved after hearing it. Preoccupied with stimuLus-response-reinforcement schedules and the like, they naturally reformulated the problem of meaning in their own theoretical terms. A referent (object) stimulates us to respond in certain ways. One is to utter a word that has, through conditioning, become associated with that ncfcrcnL. As the word becomes attached to the referent through conditioning, we transfer our responses from the referent to the word. Thus when we hear or read the word, we respond to it as we wouid to the referent. BUL obviously, when anyone utters the words red-ko: paker, none of us claps his ears in pain. So behaviorisis had to quality this simpleminded stimulus-response-transference theory to say that we respond with a partial readiness to respond, selecting out only certain aspects of the red - ho t-pufcer-experience to associate with the word and respond to. (156,157,191,199,207) Unfortunately, even if this does accurately describe what happens when we bear a word, it does not tell us very much about the meaning of Relational concepts presuppose perceptual processing. It is rot at all clear what our response- to the referents (whatever they are), of those words would be; less clear how we are supposed to be "ready" to respond to the words referring to those referents; and less clear yet how we are supposed to be ready to respond to the words in the context of a sentence. How is it that we know that when someone says something as simple as The door is open he might intend us to create in our minds the meaning I-iocked-the-door-when-I-left-sosomeone-has-opened-i t-an d-we-must-be-ca refu 17 7.1: Among the ways psychologists have tried to measure meaning objectively arc Lhese: (1) Word association. A speaker gives all the words a particular word makes him think of. (2) Sorting. A speaker puts Logether words that seem to go logether. Then he subdivides them again and again. Or he groups the groups into higher onicr categories:

136

WORDS AND MEANINGS

(3) Semantic differential. A speaker Is asked to rank a word along scales that link several pairs of polar adjectives: warm _. ,_ .cold, herd soft, fast stow, and so on. Given this admittedly incomplete description of the three theories, what is wrong with each of them ? (3S, J 57,203) At about the same time that the stimulus-response psychologists were working out their behavioral theories, linguists were trying to identify the smallest observable units of linguistic meaning. And impressed with the apparent successes of their fellow social scientists, the linguists adopted the same objectivist, empiricist stance in their analysis ^Language. Asfced the meaning of a word, a linguist would be likely to answer with an empirical definition based on what he could see and categorize, a definition that in more recent years sounded something like The meaning of a word is the sum of Us linguistic environments. That is, Lhe meaning of a word is the sum of all the sentences the word might appear in. (84,156,20Й, 233) PROBLEM 7,2; What is wrong with this definition ? And at about the same time, philosophers were also wrestling with the problem of meaning, equally unsuccessfully. The definitions ranged from meaning being what is necessary to make a proposition true, to what the words refer to, to how a word is used, to the method of proving the truth of an assertion, to the ^picture " a sentence makes. (19 J) But the principal weakness in all these definitions is that none of them allows us to deal with the specific meaning of a word in the context of a specific sentence and how ibaL specific meaning might become a new meaning. Nor do they give us a useful way to represent a meaning in formal terms. Claiming that the meaning of book is the sum of its linguistic environments or the way it is used or the words elicited by it in free association yields a definition that no one can do anything with. Indeed, it yields no definition at all. We have already suggested in Chapter Two the outlines of a theory of meaning we shall use to describe how meaning changes. We shall define the meaning of a word as the sum of the elemental components of meaning that we abstract from all the experiences we necessarily associate with the use of a word. On the one hand, it is easy enough to say now that the meaning of boy is the sum of [+object, +• count, + animate, + human, +male, -adult]

because these are the common characteristics of virtually all t h e / \ 's we have experienced. But the same criticisms we level against word-associationists or semantic differentials or word-sorters might appear to be valid here: What is the meaning of [H human] or [—adult]? Must we define the components of meaning in order to have a legitimate theory of meaning?

SEMANTIC: СНАМЗЕ

157

On the olher hand, we have ihe example of the rest of the scientific and philosophical world. There are, in every constructed system of knowledge, undefined primitives. Euclid's Elements has as its first definition Л point is [ha! which has impart. But what is the definition of par! or of no or of has? Jn number theory, not every1 eoncept can be defined. What is the meaning of "more t h a n " ? What is the meaning of " o n e " ? Tn short, we may have to be ready to accept the idea that some of the elements, perhaps a relatively few components of meaning, are simply to be accepced as undefined primitives in the system, exactly as some mathematicians accept the idea of number as an undefined pri mi liveBut even if we do this, we are faced with an equally difficult question: Given these undefined semantic components, how do they relate to one another and how should those relationships be represented? Several ways have been proposed. One resembles the model of syntactic relationships. Л fairly simple word like girl, for example, can be represented as a series of modifying structures providing more and more specific details. That is, the meaning of the word girl is a noun that is concrete, animate, human, not male, and not adult, A more complex word like confess has a more complex structure. That is, one meaning of confess is roughly rcprescntable as the diagram on page 159. Human у states to human 2 that t acknowledges that I is responsible for an action which 2 presumably did not know that 1 performed and which 1 believes 1 should not have previously revealed. This omits a good deal of secondary information such as the internal structure of state, acknowledge, responsiblefor^adion, and soon. PROBLEM 7.3: In a very general way, without relying on the very formal tree structure in the examples, sketch the semantic structure for these words: admit, acknowledge, deny, brag, criticize, accuse, b fame, praise. For our purposes here, however, we cannot possibly represent every word we diseuss in such detail, even though describing the precise nature of a semantic change on trees such as these would reveal a good deal about historical processes. The components will instead be merely listed after eaeh word in a purely suggestive way, as only very informal outlines of what kinds of components are relevant to certain historical changes. Eventually, perhaps, semanticists and psycholinguists will discover ultimate particles of meaning, the true primitives in the system. At the moment, however, we are in the position of yesterday's physicist. We know there are molecules (like man) made up of atoms [+object, + animate, 4-human, +male, + adult] organized into structures. We also know that perhaps these atoms are made up of smaller but rather obscure particles organized in ways that are not at all clear. (29, 101,102, 143,237)

SENTENCE SUBJECT

NOUN

VERB PHRASE

VERB

NOUN PHRASE NOUN

Д,

be

4,

SUBJECT

VERB PHRASE V**

ADJECTIVE

••:

be A rough paraphrase:

[ -i- concrete]

VP

VP

I

f

ADJ

ADJ

[+ani]nate]

I+human]

A noun is some noun that is concrete, animate, human, not male, not adult.

VP

I ADJ [-malcj

VP

ADJ i

I

[-adult)

SENTENCE VERB PHRASE

SUEJJLCl I NOUN

RB

[human,]

[itatei]

PREPOSITIONAL PHRASE P'

NOUN PHRASE

to

NOUN I [humanj

I

SENTENCE SUBJECT

VLRB PHRA5F

NOL'N V I I У [human,] [acknowledge] SUBJ

[should have acknowledgtJ] for

NP

[acliun,]

A rough paraphrase: Human, slaKs li> humanj lhal he acknoTvlcd£cs thai he is hjsponiibk fur an action, which h i d i d not Ino* 1 h| wasrespansLhlcfor and which h| docs not believe h] should have atVnowledjed.

160

WOHDS AND MEANINGS

WORDS, MEANINGS, AND THOUGHTS It is easy to assume that we can abstract from our experience components like concreteness, a inmate ness, and so on, because that is what reality out there is like. Words refer to areas of experience that have a natural discreteness and independence in objective rcaliLy. But if for a moment we can imagine our total perceptual and cognitive experiences as a finely meshed screen of perceptions and responses and potential ideas, then we will see (bat the words of our language cover only a relatively few areas on that screen, and that it is possible to divide that screen into many kinds of areas and subareas. There are ways of structuring areas of our experience for which we have no words, no single words, but for which other cultures do. We have brother and sister to subcategorize siblings, but Hungarian has four: batye, for older brother; tics, for younger brother; ne'ne, older sister; and hug. younger sister. Malay, on the other hand, has one word that covers not only siblings but cousins as well: saudar a. We have г о single word to cover both aunt and uncle as we have a- single word, parents, to cover father and mother, and children to cover both son and daughter. Nor do we have words to distinguish male and female cousins as we do to distinguish male and female siblings: brother and sister. Some languages have basic color terms for only black and white; others for only black, white, and red; others for only black, white, red, and either yellow or green. Some languages have as many as eleven basic, поп metaphorical color terms. In some languages, a single color term covers an area of the spectrum we would refer to with three words: red-orange-yelfow. (9) PROBLEM 7.4: Here are the great majority of OE words, including compounds, that refer to color. How does this semantic field differ from the color field of ModE? Keep in mind that the English translations cannot be considered entirely reliable. hwit (white), msfehwtt (pure white), snawhwii (snow white), sdr (white, bright), beorht (bright), teoht (bright), searohwit (brilliant whiteness), torlit (clearness, brightness), leoma (radiance, gleam, glare), scima (brightness), bkobrygd (scintillation), gt&d(shining), gizm (gleam), hpdor (clearness, brightness), hlutor (bright), blicon (to glitter, shine), seinan (to shine), bierhtan (to brighten), lyman (to shine), stimian (to shine), scimermn (to shimmer), tytan (to shine), spirting (sparkling). geolu (yellow), scilj'ar (yellow, golden, glittering), goldbleoh (golden hued), crog (saffron hued), geofuread (reddish gold), read (red, gold), rvdu (red),

SEMANTIC CHANGE

161

heoljread fbatfred), bocread (verm i Hi on), wealhbaso (foreign red, vermillior), Ыхкхшн (Sight blue), grene (green), ksewengrene(atuk&a), part! (purple), purpure (purple), basit (purple, scarlet, crimson), brunbasu (brownish-purple), bleoread (purple-), reodbasu (reddish purple), wyrtoread (purple, scarlet), weofocbnsu (whelkred-purple), weoiocread (whelkrcd-scarlet, purple), hxwen (purple, blue, azure, green'), ftxwm&nged (mixed purple), wmdeti (purple, bluish), tea/or (purple, red, vermillion), cynewaSen (of royal pwpltf^xcdeuh (fisbdye-purple). /те/и (yellow, tawny, grey, djsky, dark), dunn (brown, dingy), haw (dusky, grey ashen), healfhwil (halfwhite), grgrg (grey), hur (hoary, grey), dearcegrpg (dark grey), brun {dart, dusky, metallic, lustre), dungr&g (dark, dusky), shiwness (pallor), wtinn (dark, dusky, livid), brmwarm (dusky), sscfealu (ashyhued), aescrxg (a shy grey), cissedun (ass-colored), dndergrey). bine (bright, pallid, wan), blat (livid, pale, wan), mires (murky, dark), Ыж (black, dark), deorc (dark), dimm (dark), hodma (darkness), hserfnsweart (raven-blact), peas tor (dark), dimhiw (of dark color), xlmyrca (entirely black), brigd (play of color), bregdtm (change of color), cyiu (spotted), /ah (dappled, stained). PROBLEM 7.5: How has ModE expanded its vocabulary of color words? What have been the sources of such color words as orange, pink, chestnut, olive, violet, lavender, starlet, vermiilion, carmine, maroon, magenta, puce, azure, indigo, auburn, hazel, fawn, russet, chtmelate, tan, emerald, chartreuse, rose, cherry, plum, salmon, citrine, lemon, navy, ochre, apricot? What area of ihe spectrum had the richest vocabulary in OE4? What hue area of the ModE spectrum has the richest vocabulary? The poorest4.' Where will we find new color ten™ ? Predict one for each of the following areas: red, orange, yellow, green, blue.

SEMANTIC STRUCTURE But if we cannot always be sure Lhat ail languages divide experience into the same categories, yet the way they assign words to experience is similar enough

162

WORDS AND MEANINGS

from language to language to suggest that some semantic universal may characterise all human languages, perhaps even human cognition itself. All languages, for example, distinguish nouns that categorize abstract from concrete evperience; animate from inanimate referents; human from nonhuman; male from female. Individual items might be classified differently from language to language, but all languages systematically classify experience along these and other lines. All languages also have ways to classify objects and experiences into higher and lower categories. Animal names the class to which fish, birds, and other creatures belong. Bird, in turn, names the set to which belong the subsets pigeon, wgte, ostrich, chicken, and duck. And in turn, duck, names the set referred to as cemvusbeick, mallard, and so on. Beyond this, we have to form subsets periphrastically, by making up complex phrases: the mallards In my pond, that mallardflying pest, and so on. For these subsets, our language provides us with no individual word or term because they are not important enough in our culture to require a name. Thus, when words communicate meanings, they name not individual experiences, but classes of experiences. The more we assign experiences to the categories our language gives us, the more we refine and abstract the characteristics of those categories. The linguistic meaning of a word is the set of abstracted characteristics (which we shall call semantic features) necessary to distinguish the category which the word names from all other categories. Thus one minimal rneaning of girl must include [+young] and [—male] to distinguish the class that girl names from the classes that woman and boy name. Words for more general categories require fewer semantic features than words referring to more specific categories. The meaning of tree is composed of the features we might call [ + concrete, +Iive, -animate, + vegetative, + trunked]. The category named by pine would add at least one more feature, perhaps something like [-leafed]. Plant, on the other hand, would have one fewer feature, lacking [ + trunked]. (All these features, be aware, are purely ad hoc inventions. A thorough investigation of the semantics of plant names in English is a book-length study itself.) Certain features also imply other more general features. Anything [+animate], for example, must also be f + live] and [+concrete]. Anything [-bmale] must be [-Hive]. But some features cut across categories. Although [±malc] may imply f+live], it dc-es not necessarily imply [+human] because [ + male] also distinguishes [—human] words like bull-cow and boarsow. PRORIJEM7.6: Analyze the prefixes in Problem 6.7 into semantic features. Hyper- and sub-, for example, must differ by at least something we can approximate as [±abovc], pre- and post- by [±beforcj. How do the OH

SEMANTIC CHANfiE

163

prefixes differ in their semantic makeup from borrowed prefixes? Can features describe colors? So far, our description of meaning has emphasized logical categories. But we respond LO objects and experiences not only rationally and logically but emotionally as well. Compare the emotional responses to the words_/?//A and Jove, Unfortunately, it is very difficult to do mueh more lhan categorize these responses along a scale of favorable-unfavorable. Nonetheless, [ +favorable] may be a legitimate characteristic of a linguistic experience because we distinguish words by it. Compare do-gooder and philanthropist, intellectual and egg-head, promiscuity and swinging. Related to this aspect of meaning is another we have already touched on. We have many pairs of words which differ by social level: My instructor postponed the short unannounced examine! ion. My teacher put off ihe pop-quiz. We classify the same experience in different categories named by words that reflect how we relate to our context and to our audience. In this sense, meaning inheres not just in the relationship between the word and the referent isolated from its context, buL in the total relationship betwee-n word, referent, social context, speaker, and his audience. The meaning of rip-off is different from the meaning of steal precisely in the different ways the speaker relates to his audience and his subject at the moment he is speaking. Yet another kind of meaning that can only be defined by the grammar of the language is grammatical meaning. What for example, is the meaning of do in He does not need it', of to in He wants to fence; of o/in Some of the people left; of be in He was being helped; of have in He has gone. These are grammatical words, function words, that indicate grammatical rather than semantic information. Some words hover between grammatical and semantic information: a in a boy left is grammatical in that it indicates a noun that can be counted (Contrast a boy and *a chaos) and thai has not been mentioned in the previous discourse. But it also indieaLes singleness, a concept which is both semantic and grammatical. A truly global theory of meaning, however, would move into even more amorphously defined areas of meaning. In addition to using words, we also communicate through OUT tone of voice, the speed we speak, our gestures, our posture, even how close we stand to the person we are speaking to, AH these features overlay the words we speak to communicate messages about our emotions and attitudes. But while paralinguistics (tone of voice and so on) and kineiki (posture, gestures, and so on) are probably as systematic as the more formally denned areas of grammar and phonology, they are much less well described and certainly beyond our ability to re-create for speakers who lived before the phonograph and the movie camera.

164

WORDS AND MEANINGS

THE RECOVERY OF MEANING Complicated as all this may seem, it is part of the semantic competence of a native speaker of English. But recovering even a fraction of this information from the past h difficult, to say the least. Since we have no native speakers of Old от Middle English available, we have to rely on indirect ways to recover meaning. And from the outset, we have to acknowledge that we will recover only a narrow range of the global meaning of any given message. We will have nothing to say about intonation, relatively little to say about the way particular words might have meant different things in different contexts. We have already touched on the social prestige of French and Latin words. Given the probable atliLude of Anglo-Saxons toward kinship and war, we can make some reasonable guesses about Lhe affective meaning of a word like broSor (brother) or faederencnosi (father's kin), guff (battle), or sweord (sword). For the most part, though, we shall restrict ourselves to the most traditional areas of meaning of individual words and how their meanings have changed.

Dictionaries The first and most obvious way to recover meaning would be to refer to old dictionaries. Unfortunately, the first dictionaries of English (excluding glosses on Latin and French words and bilingual dictionaries) did not appear until the beginning of the seventeenth eentury. 7,7: Listed below are the first fifteen words from the letters: do- in six early dieLionaries. What kinds of concerns do these dictionaries appear to have had? For Minsheu, Phillips, and Bailey, the number of words from the fifteenth word to the entry for dog is also listed in order Lo suggest the degree of completeness each lexicographer strived far. Comment. Compare the same span in a modern dictionary. John Bullokar. English Expositor (1614). Docility, double, docrbjlity, dock, docket, doctoral, doctrinal, document, dodrantal, dogmatical, dollar, dolorous, dolphin, doom, doomsday-book, John Minsheu. Ductur ш Lingiias (1617). Do, doal, doater, (hating, dobekicki, doheter, doblet, doced, docibh, docilitie, docks\ docke, docked, dockes {plus 15 more- to dogge). Henry Coekeram. H i t English nictmnarie (1623). Dodec.haedron, doable, docility, document, dogdaies, dogmatical!, dogmatist, dole, dolefuit, dolorificall, dolorous, damabie, domeaticke, domesticall, domineers.

SRMAKT7C CHANGE

165

Thomas Blount. Glossographia (165ft). Docket, docibSe, docibility, docilize, doctiloqueiit, document, dodecatciyiorie, dodechadron, Dadona, dodrantal, dog-days', Doge, dagdraw, dogmatical, dogmatist. Edward Phillips. New World of Words (1658), Dobeler, dobtm, doced, docility, dock, docket, doer oral, document, dodded, dodder, dodecoedrie, dodecagon, dodecatemorie, dadkin, Dodona (plus three more to dogdoies). Nathan Bailey, An Universal Etymological English Dictionary (1721). Dobuni, docible, docile, docility, dodlize, dock, dock, dry dock, wet dock, to dock, to dock, dock, docked, docket, docket (plus 27 more to dog). PROBLEM 7,8: Befow are entries for explode from several early diction aries. What information can we recover from each entry? What appears to be the intentions of each dictionary? What information is provided ? What information is not provided ? 1. 2. 3. 4. 5. s' 7.

Bu lloker. F.x plode: to drive out wi th clap pi ng the hands, M in she w. (no e n try for explode) Cockera m. E xplodc: to d ri ve out with el apping of ha nds. Blount, Explode [explodo] publikcly to disgrace, or drive out by hissing or clapping of hands. Phillips, (no entry for explode, but under explosion) explosion, [Lat.] an exploding, a sleigh ting, or hissing off from the StageBailey. To EXPLODE [F.xplodere, L.] to decry or cry down; to mistake absolutely. Samuel Johnson A Dictionary of The English Language (1755). To EXPLODE, v.a. [explodo. LiiLin.] ^1) To drive out disgracefully with some noise of contempt; to treat with open contempt; to treat not only with neglect, but open disdain or scorn. "Him old and young / Exploded, and has seiz'd with violent hands, / Had not a cloud descending snatchd him thence / Unseen amid' the throng," Miiton. "Thus was th'applause they meant / Tum'd to exploding hiss, triumph to shame, / Cast on themselves from their own mouths." Milton. "Old age explodes all but morality." Roscom. "There is pretended, that a magnetical globe or terrclla, being placed upon its poles, would have a constant rotation but this is commonly exploded, as being against ail experience." Wilkins. "Shall that man pass for a proficient in Chrisi's school, who would have 14 been exploded in the school of Z e n o ? " South. Provided that no word, which a society shall give a sanction to, be antiquated and exploded, they may receive whatever new ones they shall find occasion for." Swift. (2) To drive out with noise and violence." But late the kindled powder did explode /The massy ball, and the brass tube unload." Blackmore. is

8.

в

William Kenrick. New Dictionary (1773). To EXPLO'DE -LX-PLODEINote: Kenrick used the numbers to indicate pronunciation] v.a.

166

WORDS AND MEANINGS

[ex plod o, Latin,] To drive out disgracefully with some noise of contempt; to treat with open contempt to treat not only with neglect, but open disdain or scorn,—To drive out with noise and violence, 9. John Ash. The Mew and Complete Dictionary of the English Language (1775). Explo'de [v.t. from the Lat ex out of, and plodo to clap the hands] To drive out with disgrace, to reject with contempt, to reject with noise and violence. 10. Noah Webster. Compendious Dictionary of the English Language (1806). EX-PLODE', v.t. (1) To decry or reject with noise; to express disapprobation of, with noise or marks of contempt. (2) To reject with any marks of disapprobation or disdain; to treat with contempt, and drive from notice; to drive into disrepute; or, in general, to condemn; to reject; to cry down. (3) To drive out with violence and noise. Look up explode in the unabridged Oxford Fnglish Dictionary, in Webster's Third New International, and in any modern desk dictionary. Compare these ModE meanings with those of the early dictionaries. 7.9: Samuel Johnson note* after each of these words that it is "low" or "ludicrous" or "cant." Comment. coax, budge, sensible (meaning reasonable), gambler, tfever,fifft, belabor, cheery, doings, dumfoutid, ignoramus, job, shabby, slim, Hark, visage, tiny, volunteer, width, hanker, fuss, scrape, squabble, Jiff, touchy, bang. What do modern dictionaries have 1<J say about these. words? PROBLEM

All dictionaries, however, particularly earlier ones, have severe limitations. By and large, they record the meanings assigned to a word by jpper-middte class speakers. Dr. Johnson's dictionary and a few others are exceptions. Only recently have those two master obscenities of English, fuck and cunt, been included in any of the modern standard reference dictionaries. (We might note, however, that fuck and cum did appear in Baile-y.) Many dictionaries now regularly include dialectal, slang, and some indecent words. But one has only to think of any piece- of current slang—spaced out, for example— to realize that in $ome ways, dictionaries are always out of date. And if modern dictionaries cannot include all the meanings of a word used by upper and lower classes, older dictionaries limited to "hard" words, as they were called, will help even less. It was not until Bailey attempted to create an etymological dictionary in 1721 that a lexicographer felt any reason to include all the words of Lhe language. So we cannot rely on older dictionaries for more than a fraction of what we want to know, both because the earliest dictionaries go back no further than the beginning of the- seventeenth century and because they dn not include all Lhe information about the vocabulary1 that we would like. A

SEMANTIC CHANGE

16?

handful of brief glossaries and collections of thieves1 card were assembled from the end of the sixteenth century on, but ever they are far from adequate. Consequently, we must turn to more indirect means for reconstructing meaning.

Reconstruction A more important source of information about the distant history of a word is in the common core of meaning retained by cognate words from diflferent languages. All these words, for example, art cognate: From Latin: Magna, as in \fttgna Carta, magnate, magnitude, magnifier, magnificence; from French, maistre, which gives English masier7 mister, and by analogy, mistress; from Germanic, much. Given these forms and others and their meanings, we can triangulate back to a hypothetical IE root, *mag~, which very likely meant someLhing close to big у г large. As a single example of how components of meaning can endure through time, here is a collection of cognate words from a variety of Indo-European languages, all of which descend from the IE root, *kel-, a root that is at least 500U years old. Sanskrit: c'5ta,a house or shed Greek: kafia, hut; kulon, eyelid; Calypso. Latin: calyx (f. Gk.) seed-vessel; calix, a cup; clam, secretly; * secret; ciliurn, eyelid; stipercilium, eyebrow; Шотк, disdainful; eelare, to conceal; cells, EI storeroom; cellar шт, a larder; cellarius, a butler; occulere, to hide; occuitaUo, a concealment; cucullus, a cowl; color, color, Romance; Old French; chalice, a cup; cale, a kind of cap. Italian: cellario, cellar; cuculia, cowj, French: amceler, conceal, Provence: cella, cell. Germanic: Gothic: huljan* cover; helms, helmet. Old Norse: huistr, holster, Scottish ; hawk, dig out. Anglo-Saxon: Jtelan, cover; liealle, hall; helm, helmet; hoi, hollow; lid, hell; hulu, hull; hold (as in hold of a ship). Celtic: Old Irish: i-ealt, cell. Welsh: cdu, ш hide. Among the words we have borrowed with the same root include calyx, Calypso, apocalypse, clandestine, cilia, supercilious., cellule, occult, chalice, cell, conceal, cellar, cowl, helmet, halberd, color, caul, and the Kit- (church) in Kilpatrick, Kitkeimy, Kilchrist, and so on.

168

WOKDS AND MEANINGS

7.10: *kel- was originally a verbal root. What meaning would you assign lo it, based on the cognate words? PROBLEM

A roughly analogous technique for recovering meaning within a contemporary Language is LO examine the variant meanings of a single currenL word. The noun play, for example, now primarily means active diversion or recreation, or a theatrical performance or its script. But there are other meanings: the play of light on the waLer, the play (looseness) in a wheel, foul play, sword play. Merely from this information, we cannot be certain that the most general meaning (movement or action) came from the more specific (recreational activity) or vice versa. But from fixed phrases like sword pfay, gun play, and the specific sense of play of light on the water, we might suspect that the more general meaning, quick movement, is the older one. When we find relatively fixed phrases like play of light or ploy in a wheel, they are often the older meanings left behind, frozen in specific contexts, while the more common meaning is a more recent development. 7.11: Here are pairs of phrases, one i I lustra Ling an older meaning, the oLher a more recent. Assuming that older meanings are often restricted to a narrow range of set phrases or special senses and that newer meanings can occur more freely, decide which phrase contains the older meaning and what that meaning was, PROBLEM

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

We assumed he was innocent. The Virgin Mary was assumed into heaven, Police try to catch croaks. The shepherd's crook was in the barn. . . . deliver us from evil, Defiiw this pack age. Get right with God for Doomsday is coining. You are {loomed, so give up, The outlines were veryfaint. Fainj heart ne~er won fair maid, There is a gliosJ in the house. Father, Son, and Holy Ghost. It's an ///-wind that blows no good. T became ill from the food. W e m u st A ; ^ the Sa bba t h. You can A:ee/t my m oney. I'm mad&t you. He belongs in the madhouse. You must leave on pain of death.) have a pain in my leg. You an a pest, Pesth ouses a re a wfu L places. My friend is a chief petty officer. Don't bother me about petty things. The flagship struck its pennant in surrender. The union struck the company. 14. Well trip the light fantastic. We took a trip to Mexico.

Citations The most reliable way to discover the earlier meanings of a word, though, is to collect and study example after example of a word in its linguistic

SEMANTIC CHANGE

169

context. For example, consider the meaning of dear (the ancestor of ModE in these senLcnces: c. с. с с

950: 1000: 1200: 1200:

"Se carnal t>;et micladear" (The camel that large *й?яг) " Unicorn is, anhyrne deor'* (Unicorn, one-horned deor) " Shep is. . . stille der" (Sheep h . .. quiet der) " Lambiss soffte & stille decr" (Lamb is soft and quiet *uw)

And consider these compounds: deorfaid (fatd-toW), deomeu (nelt-net), (hege-ftsaaf)f deorcynn (cyan-kind) , deorgear (ge at-gate), deormod , soul) and phrases like mere deor (mere-sea) and sx deor (&s-sea). It would be odd if all these references were limited to the Cerndae. And when we find citations such as these Ratons and myse and sochs male dene . . . ^vas hys mele Rats and mice and such small deer . . . was his meat it is certain that deor originally meant something close to animal or beast, and that our modem meaning of deer results from a narrowing of deor's original semantic rangeWhen we combine Lhese ways of working back into the semantic history of deer with the way Latin words meaning animai were translated into OE and with the meaning of cognate words from other languages (Old Frisian dinar, Old High German tior. Modern German tier, and Gothic dins ali mean animal or beast), we can be fairly certain that the deor of OE and ME named a more general category than our modern deer. (142) PROBLEM 7Л2: What can you tell about the meaning of these words in bold face from their occurrence in the following citations. How have the meanings changed V 1.

2.

3.

4.

Ye know the house of Stephanas, that it is Lhe first fruits of Achaia, and that they have addicted themselves to the ministry of the saints (1 Corinthians, 16.15), And well beseems all the knights of noble name / That covet in th'immortal bonk of fame ,' To be eternized (Edmund Spenser, The Faerie Queen*). Better be with the dead / Whom we, to gain our peace, have sent to pence / Than on the torture of the mind to lie / In restless ecstacy (Macbeth J Hi i). But with an angry wafture of your hand / Gave sign for me to leave you: So I did; / . . . Hoping it was but an effect of humour, / Which sometime bath his hour with every man (Julius Caesar, H.i).

176

5. fi, 1. *.

WORDS AND

He shall reward evil unLo mine enemies (Psalms 54.5). . . . Irish ladies of Strict virtue and many northern lasses of the same predicament (Henry Fielding, Tom Junes). They love to retaliate kindnesses, and hate any should think they are of a churlish nature (Thomas Tryon, The Way to I Itallh j. . . . with the most beautiful stench [of Paradise]. . .the holy stench. . . , , ., delightful stench . . . (translated from "The Dream of the Rood," an OE poem).

TYPES OF SKMAiNTIC CHANGE

Schemes for describing binds of semantic change are as numerous as those who have investigated the problem. The categories number from three up into the dozens. Part of the problem depends on how we initially approach the question. Should the principal categories arise out of the reasons for meaning change, out of Lhe mechanism of the change, or from the semantic consequences of the change? "["he theory of meaning we have adopted here—of meaning being represented as a structured configuration of semantic components—leads us to describe change of meaning in those terms. Our categories will depend on how those structured configurations change when a new meaning is associated with a linguistic form- On these grounds, we can set up four basic categories of semantic change. The first is narrowing: A word originally naming a larger category changes to name only a subcategory of the original category: deer (animal) > deer (family Cervidae). The second is widening, the reverse of narrowing: go (walk) > go (any intended or directed movement or operation). The third is traditionally called metaphor: The Form of a word naming a category with a particular characteristic is transferred to name a seemingly very different category, but one which shares that specific characteristic; grasp (enclose in Lhc hand) > grasp (understand). The fourth kind of semantic change we shall call shift: A form naming a category related by contiguity to a complex of items shifts to a discrete identifiable element in the complex: bureau (coarse woolen cloth over a desk) > bureau (the desk itself) > bureau (the organization that uses such desks).

Narrowing Г Г we list the major semantic features that roughly define each of the following words and invent some others to account in a. shorthand way for

SEMANTIC CHANGE

171

those aspects of meaning which are rather complex, we can represent what happens when a word like ekvr narrows from meaning any animal to meaning just the family CerWdae. (We shall assume that all the more general features following the entry are implied in the semantic descriptions.) Avar (deer)

liquor

mete (meat)

stol (stool)

[+concrete] [+concrete] [ 4- concrete] [-concrete] [ + concrete] [4- animate] [-animate] [-animate] [4- state] [-animate] i-hum.au] [-solid] [+solid] [4-animate] [4-solid] [4-fluid] [4-consumable] [-favorable] [+ fabricated] [4-for sitting on , . r ] That is. deor was any non-human animate entity; liquor was any liquid; mete was any solid consumable; disease was any unfavorable state; stol was any fabricated object for one person to sit on. For a variety of social and cultural reasons which we shall not touch on, speakers of English added to each of these bundles of components at least one more feature or complex of features as they jscd the word to refer to a specific member, to a specific kind of the original larger set, thereby narrowing the range of reference: deer [] [•••]

liquor

meat

disease

stool

[•--]

[-.-I

[•••]

[4-mammal] [ 4-consumable] [+flesh] [+systemic] [-arms, back [+Cervidae] (later) [—accident] + visible legs] [4- alcoholic] [4-spiritous] (Be reminded again that these are inexact and unstructured features.) PROBLEM 7.13: Here arc some words with older meanings in parentheses. In a very general w a y s suggest the kinds of features you would add to give the words their modern sense: fowl (any bird), shroud (an article of clothing), thank (from the general word for think), i-кл (a flaw), corn (any grain), hound (any dog), artillery (any large implement of war) In these examples, the narrowing has virtually eliminated the original, more general meaning. But other words radiate meanings: A new meaning does not completely replace the older one. Instead, several lines of meaning can develop from a single form. The problem again is whether a new sense means a new word, homonymous with the word that spawned the new meaning. Execute\ for example, is a form (or a word, depending on whether

172

Wf.lHns At-ID MEANINGS

we decide each different meaning requires us to speak of a separate word) that retains the original general meaning of performing any determined action, even though another nieaning also developed, a more narrow one referring to the performance of one particular kind of action: to execute a death sentence, and hence by shift, to execute a man.

PROBLEM 7.14: A question about a word like execute is whether we are dealing with one word with different meanings or with homonynious words, each with a different meaning. Given the two methods of dictionary-making we have explored so far—the traditional theory illustrated in all the standard dictionaries where several etymologically related definitions are listed after a single entry, and a theory that lists the semantic components represented by a word, how would we decide whether we are dealing with one word with many meanings, or with many words, each with one meaning, all of the same form? What difference does it make ?

7.15: Here is a longer list of words whieh have narrowed. Can you generalize at all abouL the kinds of words that might predictably narrow? About the way in which they might narrow? (In some cases, an earlier and tater meaning is provided, separated by > . The ModE meaning is not provided for those words, only an earlier meaning.) PROBLEM

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

admonish (advise) affection (the act of being affected > any affection of the mind) arrest (stop) argue (make clear) accost (come alongside in a boat > to approach anyone) addict (someone who devotes himself to anything) accident (an event) carp (talk) cunning (knowledge, skill) condemn (pass sentence) censure (judge) denizen (a citizen of a country or city) damn (pass sentence) deserts (whatever one deserves, good or bad) doom (judge) effigy (any likeness) ecstasy (beside oneself with any strong emotion: fear, joy, pain) erotic (relating to love) esteem (put a value on, good or bad) fame (report, rumor)

SEMANTIC СНА1ЧОЕ

173

21. 22. 23. 24. ZS. 26. 27. 28. 2*. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

Jjfrb(d\rt) fortune (chance) ^ftW (the enemy) grumble (murmer, make low sounds) g/pojJ (spirit) immoral (nox customary) lust (desi re i n general) /ее/- (look obi iq uely out of the side of theeye) manure (v., hold land > to cultivate land) то/es/ (trouble or annoy) oi^y (secret observances) oAw (anything perceptible to the sense of smell) pill (any medicinal ball) /Jre<#came4ffanУ situation) proposition (a statement set forth for discussion) peculiar (belonging to or characteristic of an individual) praise (from (appraise : set a value on, good or bad) reek (smoke from burning matter > produce any vapor) retaliate (repay for anything) scheme (horoscope > diagram > plan) success (any outcome) stink (any odor) starve (die) seduce (persuade someone to desert his duty) smug (trim, neat) syndicate(agroupofcivilauthorities > anygruupofbusinessmenpursuing a common cornmercial activity) 47. smirk (smile) 48. suggestive (that which suggests something) 49. sanctimonious (holy, sacred) Does any of these words retain its earlier meaning? PROBLEM 7.16; I llustrate how any ten ModE words might narrow. Some Causes of Narrowing Wliy words narrow, though, cannot be discovered from purely linguistic data. But we might gain a possible insight into one reason some do From the way pill has recently narrowed from its original general category of smalf medicinal ball to the more specific meaning of birth control pill. Two forces, one social, Lhc и the г psychological, have brought about the change. First, since birth control has become important in our society and therefore frequently discussed, a category of objects like pills that control birth (actually,

174

WORDS AND MEANINGS

they should be called conception control pills, but that, perhaps, is a bit too explicit) requires a name, not a phrase. Second, since the principle of least effort always operates to create the shortest convenient term for a useful meaning, the full compound phrase, birth control ptft, became shortened to The Pill. Using a theory of semantic features, we can say that the semantic components originally attached to the individual words shifted to the last word, the word that originally named the superordinate set. Schematically, the change can be represented like this: birth control pill > The Pill [+A...] [ + B . . J [+C...J [+C, + B, +A] It is as if the underlying syntactic structure of the phrase becomes the semantic structure of the word: pill controls

[someone

gives-birth-to

baby]

That is, the pill controls someone's ability to give birth to a child, a sense compressed into birth controlpitl, and even further into ThePiit. The meaning seems to have a syntactic structure. 7.17: In Chapter Six under Conversion, we discussed how adjectives can convert to nouns by the same principle of economy. How would you formally describe how the meaning of such words change? What difficulties arise? Demonstrate with these terms; criminal person, private soldier, chemical substance, vegetable plant, general officer, mortal person. PROBLEM

In other cases, the narrowing of one word may have been influenced by the fact that two words were available for roughly the same meaning A. When one of the words already has two meanings, A and B, it is often the case that it will lose one of them, say A, if the other word available for A is more prestigious. Deer and beast, for example, both meant animal in ME, but by this time, another meaning of deer was the animal we now associate with that word. Since beast, a French borrowing, was undoubtedly more socially prestigious, it probably helped force deer to lose the general meaning of animal. The same thing may have happened with spirit-ghost, odorstench, table-board, chair-stool, desire-lust. The first word in each pair is a Romance word, probably more prestigious at a time when the native word had two meanings (or more). The Romance word monopolized one of the meanings while the native word narrowed to one of its more restricted meanings. In some instances, the French word may have helped speed the disappearance of the native word altogether :j'ace-hleor, voice-stefnT peacefrW,people-leod. (188)

• SEMANTIC CHANGE

175

In other cases, however, a native or borrowed word was available to replace the original semantic space of the narrowed word. Thus instead of a word being pushed toward a narrowed meaning, the empty semantic space left by в word that had already begun to narrow created a kind of semantic vacuum that pulled another word into its space: dog replaced the narrowed hound; bird replaced the narrowed few!; smile replaced the narrowed smirk; the Danish die replaced the narrowed starve.

Widening The opposite of narrowing, logically enough, is widening. And it is obvious how a theory of semantic features would represent the simplest examples of this process. The least linguistically based kind of widening occurs when a culture adding radically new objeets or processes to the experiences of its speakers uses an old name for a new object. Picture, for example, once meant only a painted picture, then also a drawing, then a photograph, a cinematic picture, an X-ray picture, a TV picture, and now even a radio-telescope picture. Where the referent was once a fixed stable object, picture now names a class of things that can break up and reassemble themselves, dissolve, and in the case of laser holography, even be three dimensional, Saii once meant to cross water propelled by windpowcr. Now we can "sail" under the ocean in nuclear powered submarines. Pen once meant a feather used as a writing instrument. It now covers bail point pens, fiber tip pens, fountain pens, drawing pens, electrical styluses for drawing on cathode ray tubes. A barn was once a place to store barley (from bere-егп, or "barleyplace"). The word now covers almost any large agricultural building. Holidays were once "holy-days": Christmas, Easter, and so on. Now holidays include Labor Day, Veterans Day, and for school children, even the days when the teachers go out on strike. All of these meanings have ehanged because our culture has ehanged. For the most part, such changes are close to the "artifactual" meanings attached to borrowed and created words that name items of commerce, naLural flora and fauna and so on. The meanings have widened because a wider set of referents has gradually appeared to expand the semantic space the word must cover.

PROBLEM 7,IS; We can probably assume fairly safely that when words widen they lose semantic features. Bird, for example, once meant only young or small fowl. The widening to mean any feathered creature, the modern sense, can be represented like this:

176

WORDS AND

MRAN1MG5

bird l [+concrete, 4-live, 4-animate, —human, + avisr —adult]. bird ; [ •+ concrete, -Mive, + animate, — hunian, +avis] L Without trying to specify afl the features for the following words try to be moderately specific about the components lhat have been lost. Can you generafize at all about widening? About some kinds of words that widen? 1. box (a small eon lai ncr made of boxwood) 2. allude (mock) 3. for;^ (bring a bow i nt о tensio n wi th a bow string) 4. aunt (father's sister) 5. rtrunifi(tho smell of spices) 6. butcher (one w ho sia Lighters goats) 7. №rf (you ng of the family avi$) 8. carry (tra nspo rt by cart) 9. chicken (n young hen or rooster) 10. divest (remove one's clothes) 11. dki (excrement) 12. dephre (weep for) 13. detest (condem n, cu rse) 14. elope (run away from one's husband) 15. frantic (madness) 16. frenzy(wilddelirium) 17. fact (a thing done) 19. gang (Ei set of tools laid out for use > a group of workmen/slaves) 20. hun-eat (reap ripened grain) 21. holiday (a holy day) 12. jowtnsy (a day > a day's trip or day's work) 23. flif7£Jc(theknowledgeandskilloftheMagi) 24. njejjfamealsetoutforagroupoCfour) 25. mystery (divine re vea led к no wledge) 26. rt!fW(memory > thought, purpose, intention) 27. wflMjerftheinodeofhandlingsometbingbyhand) 2N, picture (a painted likeness) 29. /?/afj/(ayoungsliporcutting) 30. oil (olive oil) 31 • ordeal (trial by torture) 32. start (move suddenly) 33. scent (a n i ma I odor for t rack i ng) 34. surly (sir-[y, that behavior which characterizes a "Sir") 35. sitly (deserving of pity > frail > simple, ignorant > feebleminded) 36. sail (cross water propelled by the wind)

SEMANTIC

/77

37. stop (fiH or plug up ; prevent passage by stopping up > prevent the rnovemeut of a person) 38. sanctuary (a holy place) 39. slogan (the battle cry of Scottish clans) 40. uncle (moLhcr's brother) It is harder to find a pattern for widening than it is for narrowing. It is not entirely certain, but meanings seem to widen somewhat less frequently than (hey narrow. As a culture becomes more diversified and more complex withTrtore areas of knowledge and activity, those areas require a vocabulary. Because every language has a finite number of words and because speakers arc not inclined to coin completely new forms for new concepts, the simplest way to deal with new areas of knowledge is to use the current vocabulary. Borrowing, derivation, compounding, and so on operate here. But perhaps even more frequent is narrowing. But on the other hand, it can also be difficult to talk about the most ordinary activities of daily life as they diversify. Once it becomes possible to drive {drive originally meaning to force an animal along), or ride {ride originally meaning to go on horseback), or walk (originally meaning to travel about in public), then talking about getting some place without specifying how becomes difficult. The word go, originally meaning to walk, generalized so that ал Fnglish speaker can now say I am going to town this morning without having to specify how he sets there. Carry generalized from transporting specifically in a conveyance of some sort to transporting by bearing up in general: The wind carried the seed, and so on. Like narrowing, widening also leaves behind! relic forms. For the word sanctuary, we have bird sanctuaries, fish sanctuaries, poliLical sanctuary, the notion of safety in general But a good many Christian denominations still call the main meeting room in which the congregation gathers for general worship the sanctuary. One meaning of start is still what we do when we hear a loud sound. Mind still means memory in remind. Scent still means animal trace in on the seen!. PROBLEM 7.19: What are the relic forms for these words: fact, gang, mess,

mystery, stop! PROBLRM 7,20: Take any ten words from the list that widened in Problem 7.18 and ten from the list that narrowed in Problem 7.14. Narrow those that have widened and widen those that have narrowed so that they mean something different from what they originally meant. For sample, erotic narrowed from a general reference to !ove to mean sexual love. Conceivably, it could widen past love to mean any strongly sensual experience: erotic food, erotic and so oa.

17$

WORDS Artp MEANINGS

Transfer

A third process of semantic change usually goes by the term metaphor. To keep our Urrms consistent, though, we shall use the term transfer to parallel the other two spatial metaphors of narrowing and widening. Transfer is a kind of sideways leap of a word through semantic space. It occurs when two categories of actions or objects resemble each other in at least one feature and one of them has a name while the other does not but needs one. The name of one is then transferred to the other unnamed category of скpcrience. This kind of transfer often results in a large gup between the meaning of the original word and the new meaning. There is. for example, no immediately obvious connection between what happens when the muscles of the hand contract the fingers around an object and what happens in the mind when if understands a concept. But not later than 1680, someone who saw a connection transferred grasp from the category referring to clutching to name the category that includes the conceptual experience. The similarity is probably in the outcome of the actions (encouraged, perhaps, by the parallel history of Latin comprehend). In many cases, of course, the connection is instantly apparent. The vertical members that support the flat surface of a table Took somewhat like legs, partially function like legs, so having no peculiar name of their own, they are called legs. But in most cases, the connection is not immediately apparent. What, for example, were the original shared features behind the literal and metaphorical meanings of strung out or groovy ? And in some cases, the original sense has been entirely lost. The idea fizzled out contains a dead metaphor whose source we can recover only from dictionaries that supply complete etymologies. PROBLEM 7,21: Here is a passage of straightforward expository prose. What are the obvious metaphors? What are the borderline cases? When does a word cease being a metaphor ? Once the Nazis were beaten, many of us were glad to forget those issues [the open and covert approval of anti-Semitic activities in Germany from 19.41 1945], or leave them
SEMANTIC CHANGE

179

rising unemployment, a long tradition of fear and hate that goes back, say, to Martin Luther's later years. From kobert Coles "Understanding WFiitc Racists," The New York Review of Books, December 30, 1971. p. 12, Bat if metaphor is so common, it is hard to describe, for the process is immensely complex. No one has yet offered a theory of metaphor that satisfactorily explains how it happens, much less how it is possible to understand one. Most linguists deal with, it as a bastardized form of language, concentrating instead on what they believe to be the central problem, its ordinary, literal use. But any phenomenon that occurs as frequently as metaphor, that is so important a source of linguistic change cannot be treated as anything except central to our ability to use language. Consequently, metaphor must test any theory of meaning designed to account for how humans use and respond to words. We might initially approach metaphorical transfer something like this: GRASP

(GRASP:> 4 action — physical

I-ultimate

1..Г1

-

. I

ttomratt flexible appendages, aroun

We have a Category of Experiences 1 from which we abstract a characteristic set of defining Features; and associate with the utterance form GRASP,. The Feature*! become the meaning of GRASPi- Then around 1680 perhaps, there was for someone a distinct though unnamed Category of Experiences^ which were characterized by a set of Features^ But this Category of Experience^ and its Features, had' no Wordj to represent them. Understand, apprehend, comprehend, realise elicited, meanings that were all close, but did not quite include certain Feature^ that were crucial to Category of Experiences*. But enough of those Features^, specifically those features we shall represent as [jS], were atso among Features! abstracted from Category of Experiences!. So some speaker who consciously or unconsciously recognized the shared features \fii-2] shifted GRA£P[ from Category of F\periencesi to name the unnamed Category of Experiences; and represent the Features2. But now the problem arises of how to represent the features in common, _i. Most of the Features, were, of course, left behind. G R A S P L is physical

ISO

WORDS AND MEANINtrS

and not necessarily [-1-human!. ORASP2 is mental, limited to humans, entirely an intellectual experience. Indeed, the shared features, [fi]i-2, are so abstract and so amorphous that it is difficult even to state them prosaically. They must include something like a successful completion of an intention, the at least temporary "connection" between subject and object. But when we try to become more specific than this, we fall into yet other metaphors; encirclement in the mind, seizing an idea, fir in control, and so on. Or we chase synonyms down a path that takes us away from tiie unique meaning of GRASP2. Metaphor, in fact, often seems to highlight aspects of meaning that no dictionary tan hope to capture. Here is another example: to tsiipick. The original meaning was to pick lire-eggs from the body or clothes, an operation which, we might assume, requires a close, searching examination for tiny objects that have to be painstakingly plucked out with the fingernails, (Nitty-gritty may come from the sit me source. Nitty means filled with [ice eggs; gritty is probably a rhyming reduplication of grits, or coarsely ground grain. Thus grits filled with nits would be the infested remnants of the grain, the last and least palatable part: Now when we try to specify exactly what Features] we should include in the literal sense of NITP1CK, thai also belong to the group of Features2, features that characterize the "cavil" sense of NITPICK 3 , we discover that our original ideas about meaning may be inadequate. Let us assume that these two lists of features very roughly and very inadequately define the two words: [+action, +• physical, + animate, + human, + search to remove nits.] N1TPICK 2 : [+action, —physical, + animate, + human, + cognitive, + cavil over small details]. This, however, does not specify how they arc essentially alike. We lack the notions of the extreme tininess of the objects searched for, of their unimportance or of performing this action over an unnecessarily long period of time, of the unfavorable associations with both activities, of the judgment we make about the person who engages in such tasks. Or consider this final example. The words muscle and mussel are both inetaphorit; transfers from a word meaning " little mouse." These words presumably transferred because a muscle contracting under the skin and the crustacean in the shell reminded someone of the referent of the Latin word for mouse. But precisely what common features would we have ю create to reveal the similarity between the fibrous tissue moving under the skin and the rodent or the shellfish ? To solve this problem, we must distinguish two kinds of meaning. The

5ЕМЛ1ЧТГС CHANGE

181

Features so far chosen to attach to words have been selected (1) to predict how the word they associate with behave* in a sentence and (2) to distinguish one word from another. Thus resemble would have to be marked something like [-voluntary] or J+stativeJ because the action referred to is a state beyond voluntary control. We do not say *He was busy resembling his father, as opposed to He was busy imitating Ms father. On the other hand, mimic and ape can occur where imitate occurs: He WOS busy imitatingjapinglmimicking his father. But they are not precisely synonymous. Thus we would have to formulate some other features that would distinguish imitate from ape from mimic. But once we have postulated features to accomplish both of these goals, we have not necessarily exhausted a word's global meaning. One meaning of a word like salt, for example, could be defined as crystallized sodium chloride, a definition that could be reduced to features to distinguish it from other substances- But in addition to this basic level of meaning, we know a good deal about salt that is not necessary to distinguish it From other referents or to know how to use the word salt in a sentence. We know that it is mined, that the sea is full of it, that it is necessary to life, that it burns if it gets into a cut.. . . If we decide that all this is also part of the cognitive network associatfid with salt and that we want the word meaning to name that cognitive network, then we will have to distinguish at least two categories meaning. But while semantic features appropriately structured might conceivably describe the mininrul meaning and distribution of the word salt, that is, its linguistic meaning, such features cannot easily account for the second kind of meaning, the encyclopedic meaning of salt. TH Structure of Metaphor The answer is not to exclude encyclopedic meaning as a kind of meaning, because we have to recognize that metaphoric transfer depends usually, if not always, on our ability to incorporate the encyclopedic meaning of a word into its transferred linguistic meaning. At one time part of someone's encycloiwdic knowledge about the category named nitpicking was that literal nitpicking was boring and unpleasant; about the category named grasp that when we grasped something we often firmly controlled the object. But none of this would have been relevant to the origina! core linguistic definition of literal nitpick or grasp. In fact, the crucial aspect of gn>sf}^flliclLi transferred to graspvansun» the feature indicating some kind of "firm control over," is only occasionally associated with g r a s p ^ . ^ : Consider He grasped the tiger by the toil. By resting our theory of meaning, metaphoric transfer forces us to modify it. Apparenily, part of our linguistic ability lets us incorporate our virtually unformaliiahle encyclopedic knowledge into our more or less formalizablc

IH2

WORDS AND MEANINGS

linguistic knowledge. Schematically, the process seems to operate something like this; NITP1CK,

•* (NITP1CK 2 )

Encyclopedic meaning

Linguistic meaning

Global Meaning

Global Meaning

The—as yet—unnamed experience on the right, which we can roughly paraphrase as "cavilling over details," includes in its central core of linguistic meaning the features that would represent the allegedly unnecessary and unpleasant painstaking starch for small details. Part of this meaning, searching for something (features D and E, let us say) is also part of the linguistic meaning of literal NITPICK^ Another part, the sense of unpleasant and painstaking, is part of the encyclopedic meaning of NITP1CK 1 (features В and С let us say). Whoever wanted to express the exact quality of distaste for the unnecessary cavilling over details in a single, expressive word decided that NITPICK, had enough linguistic and encyclopedic features to qualify it as the name for the category of experiences. So he transferred it. A question which always arises in discussions of metaphor and linguistic change is how metaphor of the bind we have discussed here differs from poetic metaphor. Some have argued that poetic metaphor is more intense, more emotive, more extreme. Thus a metaphor like the following is " p o e t i c " : There might you heare her kindle her soft yoke / In the close murmur of a sparkling noyse. (Richard Crashaw, "Musick's Due!.") Rut synesthetic metaphors of this kind have characterized the development of English since the earliest

SEMANTIC CHANGE

183

of 01:. Bright sounds is a phrase whose metaphoric origins are stii! alive, but it could be used in either poetry or ordinary language. Perhaps a more basic distinction lies in the perceived intcntiun of the speaker who creates the metaphor and the context in which it occurs. If in order to express a new concept one creates a metaphor like His deep voice was atsa rather Barrow, the metaphor operates within the limited con Lex t of that senLencc. The speaker ordinarily does not intend that his audience will respond to every possible nuance of meaning in terms of (be total contest. If it somehow taps an area of semantic space the listener and enough oihiirs feel must be named, narrow might well permanently transfer from spatial reference to aural referenceBut a poet who wrote The narrow voices that escape the red-lipped j lipteix mouths thai. . . often intends thai, the metaphor be part of a larger cognitive and emotive whole, that his audience understand that behind it is a rather different intention, that his audience be sensitive Lo implications, that they develop them in ways no audience can in ordinary discourse. Thus the same apparent metaphor, a narrow voice, must be responded to in very different ways, depending on whether the audience understands it to be in the context of a poem or in ordinary language. A difference that has come to characterize much modern poetry is the degree to which some poets rely almost exclusively on private encyclopedic knowledge as the source of the similarity between two cognitive spaces: •

I am soft sift

In an hourglass—at the wall Fast, but mined with a motion, a drift, And it crowds and it combs to the fall; I steady as a water in a well, Io a poise, to a pane. But roped with, always, all the way down from the tall Keils or flanks of the voel, a vein Of the gospel proller. a pressure, a principle, Christ** gift. From Gerard Man ley Hopkins, The Wreck of the Dcutschland Whatever the shared semantic features of the metaphors are in these lines, they are not immediately obvious. 7.22: Here is a list of words, most of them with multiple meanings, all of which have at least one metaphoric sense. High, for example, in a high sound is a metaphor transferred from the literal spatial sense of high. If the metaphoric meaning is not immediately apparent, a phrase illustrating the original nonmetaphoric sense follows in parentheses, in other cases, an earlier literal meaning is preceded by/. As you inspect the list, consider the following problems: (1) Pick at random four or five different kinds of words and try to PRORLEM

184

WORDS AND MEANJNOS

state in non-metaphorical language precisely what features the original sense and the metaphorical sense share. (2) Are there any generalizations about how certain classes of words are susceptible to metaphorical transfer? (3) Arc there any metaphors that characterize certain kinds of activities, social contexts, personal relationships? Which of these metaphors are likely to be the most durable? Which the most transient? Why? Which have lost all sense of their metaphoric-af origins? Why?

1. 2. 3. 4. 5, 6. 7. S. 9. 10. 11. 12.

13. 14. 15.

1*. 17. 18. 19.

20. 21. 22,

23. 24.

25. 16. 27. 2S,

29. 30. 31. 32.

abstract (f, draw away from) advert (f. turn to) affirm (f. make firm) analysis (f. separate into parts) animal (You animal!) bright (a bright idea) bitter (sharp to the taste: He is a bitter person) brow (the brow of a hill) bewitch (a bewitching aroma) &tf?[YouoLd bat!) breed (I have no bread to spend) blast (We had a blast at the party) blow up (He blew up in anger) cold (She was cold to me) cool (Coal \t, man) conceive (f. to catch) conclude (f, to enclose) concrete (f. to grow together) conned (f. to tie together) cut (She cut me dead) crane (the derrick that resembles the bird) compose (f. to put together) comprehend (f. seize) cat (She's a cat) dig (I dig that idea) deep (deep thoughts) dark (I'm in the dark on that) define (f. place limits on) depend (f. hang from) dog-(You dog!) dirty (a dirty mind) dough (money)

33. 34, 35, 36, 37. 38. 39. 40. 41.

42. 43. 44.

45. 46. 47.

48, 49. 50. 51.

52. 53. 54. 55,

56. 57. 58. 59. «0.

61. 62. 63,

64. 65.

e's a drip) drag (This class is a drag) eye (the eye of a hurricane) exist (f. stand out) explain (f. make fiaL) enthrall (f, to make a slave of) foot (foot of a mountain) fuzzy ( Г т fuzzy headed) finger (a finger of land) fascinate (f, enchant by witchcraft) flat (a flat note) get( \ don't get it) grasp (I grasped the concept) guts (He has a lot of guts) groove (I'm in the groove) gas(]t wasagasser) grab (How does the idea grab you?) high (He got high on dope) hang (He's always hung up) heavy (I had a heavy time) hot (a hot idea) heart, (the heart of the problem) head (head of the line) hands (hands of the clock) home (drive the point home) Jazz (f. sexual activity) jice(.[ don't dig this jive) intelligent (f. bring together) keen (f. intelligent) load (a load off my mind) long (a longtime) tond(a loud color) tip (tip of the glass)

SEMANTIC CHANGE

66. 67. 68. 69. 70.

71. 72. 73. 74. 7576. 77. 78. 19. 80.

lamb (She's a lamb) тон'Л (m.o и t h of a ca ve) mouse (You're a mouse) milk (He milted the job dry) pagan (f. civilian, in distinction to Christians, who called themselves soldiers of Christ) pig (You pig) quiet (a quiet color) i'ip-njjfX 11 was a big rip-off) rough fa rough voice) f*'fo {the ribs of a ship) report (f. to carry back) result (f. to spring back) shallow (shallow ideas) soft (a soft wind) sharp (a sharp dresser)

185

81. 82. 83. 84. 85. 86. 87. 88. Я9, 90. 91. 91. 93. 94. 95.

smooth (a smooth operator) salve (Г. to break up) spirit (f. breath) .moto?(You s n a ke!) straight (He's a straight guy) square(H e's a square) split (Let's split) /jfra on (He Lurns me ou) fry> (Ft was a bad (drug) trip) thick (a voi ce th ick. wit h anger) tit in (a thi n sound) tongue (a tongue of land) translate if. carry across) warm (a warm color) wrestle (I wrestled with the problem) 96. wild (л wild idea)

These exercises should also demonstrate that metaphors can function with varying degrees of liveliness. Л completely dead metaphor is a word that is no longer a metaphor, one like result, from the Latin sallre, to leap. Resuft earlier meant to bounce hack. From the same root are salmon (the leapcr), somersault, assailant (someone who leaps on one), assault, resilient, exuft (from L. e.vsultare, to leap, ex- is an inLensifier), insult, and insolent. The specific notion of leaping is dead in many of these, but becomes immediately apparent once we know that .Wire lies behind them. Other metaphors become apparent if we think about them for a moment: sweet music, dopey people, grasp, perhaps even understand. These are the metaphors that are well on I heir way toward becoming dead metaphors. And finally, there are those transfers that instantly present themselves as metaphors, indeed, whose Jtietaphorie power depends on our recognizing them as such: rip off fox steal, cracked for crazy (which itself once meant cracked or broken, as in crazed pottery), nosey, and so on. Most of these, of course, are slang words, a kind of vocabulary we shall deal with later. FRORLEM 7.23: Take any five nouns, verbs, or adjectives and describe how they might metaphorically Lransfcr to a new meaning.

Shift Л fourth way we associate existing forms with new meanings occurs when some discrete clement within a larger structure of related or connected

1Я6

WORDS ANO MEANINGS

objects, events or operations part of either the linguistic or encyclopedic

meaning: 1. 2. 3.

takes its name from the name of the larger structure; gives its name to the larger structure; shifts to or lakes its name from another element within the larger structure.

We shall call this process semantic shift. Changes of this kind give us some of our most unusual histories of semantic charge. We have already mentioned the etymology of pedigree. The individual element, the three marks that looked like the foot of a crane, gave its name to the abstraction connected with the larger whole: the quality of descent. Hearse is another word of this kind. It first meant a triangular harrow or plow. Then it metaphorically transferred to a triangular frame that held church candles. Then it began to shift: to the framework that held candles over a colh'n while it was hcing carried into the church, to the framework that held tapestry curtain* or the pall over the tomb or coffin, then loosely to the coffin or tomb itself, then to ihe conveyance that transported the coffin (and later, metaphorically, to the conveyance that transported pianos in nineleenlh-ceniury London). This is a case where discrete elements within a larger whole required special names and the most readily available one was attached to another element within the whole. Toilet is a yet more complicated example. It began with Latin tele, a weaving or web. ll gave Old French teite then loite, a cloth, which narrowed to the doth in which clothes were wrapped. The meaning then shifted to the cloth used to protect clothes while a woman was setting her hair. The diminutive of toilet totfeite, then was attached to the little cloth cover for the dressing tabfe, then to the items on the table, then to the piece of furniture containing them, then to the room in which the object stands and in which one dresses himself, then to the action of grooming and dressing, making toilet both concrete (furniture) and abstract (dressing). Thus we should not be taken aback when we read that it was once fashionable for eighteenth-century women to entertain their callers during their toilet. Then apparently a social change occurred. Classes of speakers other than upper class used the word for the room in which they performed their dressing and grooming. But at this social level, the room had to serve more functions than just a place to dress. With the spread of indoor plumbing, the room came to include the bathtub, the washsland. and the convenience earlier known as the close-stool (cited as early as )4lu), jokes (1530), the John (c, 1650), tht с inset of ease (! 662), waterdojet (1775—now often shortened in Great Britain to ИЛС), lavatory (1845), commode (1851), and in less polite contexts, as the ran, head, etc. Since this most frequently used object required a polite name, it is not surprising th:it earlier it attracted to itself ihe name for

SEMANTIC CHANGE

187

the piece of furniture in which it was kept and then once ii achieved tie technological level of being built in as part of the room, ihe name for the mom itself. Television seems to have seized on bathroom bowl as the latest term. Thus toilet was almost certainly at first a euphemism for ihc plumbing fixture we now call that name. When toilet became too closely associated with the object (as the other names did before it), it left no name for the room containing it that many speakers could use without embarrassment So the room was re-named after ihe least frequently used object—the bath. Most of us now prefer to say we have a toilet in the bathroom (even 1Г it contains no bath) than a bath in the toilet, which would be "historically" correct but now soc ia Hy i m possi bic. Another example: The word bead, which now means the little round ball strung on a string, derives from OE gebed or bed, meaning prayer. When a bedesman says his pater noster or rosary, he keeps track of the prayer by counting the small balls. Sentences such as How many beads have you SOirf, Have you finished your beads, and so on could all be construed as referring either to the little round halls or to the prayers. Since the speakers apparently sensed a need to name the particular little round balls used in their prayers, they shifted the word for prayer, spelled bede, at the lime, to the little round balls. (The word then generalized to include all little round balls, then to all little round shiny objects: beads of weal, heads tf glass. Then it partially narrowed to the namt of the round ball at the end of a gunsighi; thence, to take a bead on someone. And that has widened again so that we can say Arnold PaMer took a bead on the PGA title today.) PROBLEM 7.24: How is the kind of shift described for bead different from widening? How is the shift of toilet different from narrowing? Narrow, widen, and transfer toitei. PROBLEM 7.25: Pick several words that name elements within a larger whole and speculate how they might shift to name unnamed elements in that whole, PROBLEM 7.26: Reconstruct the context and process by which the words ш bold face derived their meaning in these phrases: a drinking glass, а six-Хгол, a nickle candy bar, rubbers and galoshes, bed liatn, о pencil lead, an expensive iw Jockey silks, sheer nylons, a cmkfor a boirie, Woolite for your woolens, read the paper. PROBLEM 7.27: The following words are ambiguous. Explain the ambiguity and the process by which the ambiguity arose. How regular a process of word creation is this? What generalization can you make4.' report, illustration, building, discovery, finding, statement, description, explanation, account, announcement, invention, creation, production, representation, supplies, assembly.

l№

WORDS AND MEANINGS

PROBLEM 7.2S: Using the model we presented in the discussion of metaphoric transfer for the represemation of linguistic and encyclopedic rn ear ing, explain how the formal mechanism by which shift takes place differs from melaphoric transfer. That is, the linguistic meaning of beds was once that required by the meaning we now associaLe wiLh prayer. The encyclopedic meaning included everything we know about prayers—where they were said, who said them, and what was used in association with them ; little round balls. The usual terms for these shifts come from classical rhetoric: Metonymy, when the name for Lin attribute or adjunct shifts to name what it is adjunct to: king = crown; and synecdoche, when a part names a whole or a whole a part: worker = hand. Bat this distinction is irrelevant when we recognize that a single cognitive operation is occurring here. We perceive complexes of parts of groups or events as wholes, not as unstructured collections of discrete parts. Gcstalt psychology concentrates on this aspect of mental activity. We perceive and conceive of aspects of experience in a Gestalt, a figure comprised of parts in a specific relationship. When the GesLalt is significant in the culture, when it recurs in essentially the same form, we inevitably give it a name: war, graduating marriage, trials and so on. Sometimes we create the Gestalt for others by creating a name: Establishment is a word that pulls together a diverse range of elements into a single cognitive structure that may not have ''existed" in the minds of many speakers before the meaning was explained. Sometimes, of course, the Gcstalt exists but does not yet have a name. Not too long before I885t for example, the Gestalt of Lwo persons, usually of the opposite sex and unmarried, meeting at a particular lime for a shared entenaininenL, had no distinctive name, partly because the event as we now know it did not exist until relatively recently. Rendezvous, appointment, tryst, engagement, assignation, were not quite right. Since a time and date had to be set for the meeting, the element date in a phrase like We have a date Knitted from the time of the event to the event itself.

Drift There is a Kind of semantic change that at first glance may appear to constitute a fifth class: When a word naming one area of meaning drifts into an adjacent and coordinate area of meaning. For example, naughty onec meant wicked and depraved. Today, it means only mild mischief. Mischief itself once meant wicked behavior. Today, it can apply to a merely naughty child. At one time, shrewd meant depraved or wicked. (It comes from the noun shrew or a wicked and malignant man, the meaning it had before it meant a nagging woman. But before it meant wicked man, it referred to the

SEMANTIC CHANGE

189

small animal, which was believed to be a viciously evil creature.) Today, shrewd is somewhat complimentary. If we lay any of these pairs of meanings on a scale, we can sec how it is possible for a word to move from one point to the next by a scries of small steps through overlapping semantic areas:

I evil

2

4

keen-witted

We might postulate sets of components in overlapping steps that would result in the scmanLie changing from one sense to the next: 1 2 [evil] i — - и [evil} *

3 к [evil]

4 [cunning]

/ [unconscious] [conscious] [cunning] [practical] ^

5 [keen-witted] *• [practical]

Each stage shares at least one putative feature with the ne\t, allowing the word to " slide " through semantic space. But this kind of gradual change can probably be explained better by successive widenings and narrowings. For example, the last citations of the evji sense of mischief dale from the nineteenth century. The earliest citation* of the merely vexatious meaning date from at least 17&4. We can say, then, that at least by I784t the sense of mischief widened to include vexatious actions as well as evil actions, to a class of actions including both. Only the use of mtsehiefmth, say, children ot the denii would specify the exact nature of the action. Then in the nineteenth century, at least for many speakers, mischief narrowed to designate only the class of vexatious actions. The same thing probably happened with shrewd (though the number of meanings multiplied greatly, as well). 11 meant malignantly wicked from the fourteenth through the seventeenth century (with a few citations into the nineteenth). The clever,'keen-witted meaning is cited first in the sixteenth century. We mighL wonder whether during that period of overlapping reference, shrewd did not have two meanings, but rather a single wider one that could refer to both a slightly weakened sense of evil and keen-wittedness. Then in the seventeenth eenlury, shrewd apparently narrowed to exclude the wicked sense, leaving only the modern sense. PROBLEM 7.29; Here are some words, followed by their earlier meanings. Suggest how drift might explain their current meanings. 12, 3.

mischief: evil plight > harm, moody: brave > proud > angry. pretty: craft/cunning > proper appearance.

190

4. 5. 6. 7. & 9. 10. 11. 12.

wemns AND MEANINGS

grin: draw lips back in pai n or a nger, quick: living. gaudy: luxurious. Uftcouih: unknown. sly: skillful. cunning; learned. coy: quiet, crafty: strong. surly; imperious.

PROBLEM

7.30: Show how five modern words might drift in the future.

Some Causes of Drift The conditions for this kind of drift are found in the way we think and speak. Very few of our named semantic areas comprise sharply defined categories. Most blur at the edges. That is, while conceptually a triangle is a triangle is a triangle, real objects like streams tend to shade off into rivers at one extreme and into rivulets at another. This kind of indeterminacy should disturb no one who is sophisticated about language, for there will always be cases where we cannot be certain whether any particular object qualifies for membership in a particular set, The naturally blurred edges of categories and the infinite varieLy ofexperiences we have every day allow such words to drift through semantic space. Why words should drift in the specific direction they do, however, is not always easy to predict or, even after the fact, to explain. There are, however, some generalizations. Words susceptible to value judgments generally drift toward less specifically unfavorable meanings, probably because a speaker tends to exaggerate and use a word that is too strong for the situation he refers to. Through constant overstated association with a situation less extreme than the word originally referred to, the once-strong word drifts to a less strong meaning. The drift of mischief from evil to almost lovable annoyance almost certainly began when actions that were almost but not quite malignantly evil were ealfed mischief. Once mischief became attached to that lessened sense of badness, it was used once again in an exaggerated way for something [ess serious yet, perhaps for oniy mindless vandalism. Once attached to that meaning, it again was used to refer perhaps to only potentially serious fun. And once attached to that, it was used again in an exaggerated way to refer, perhaps, to a baby pulling pots and pans out of a cupboard. Almost the same change has occurred with naughty. 7.31; There is in English a class of words called intensified that illustrates this tendency to drift perhaps better than any other group of words. Intensifies all modify adjectives; in the sixteenth and seventeenth centuries they were frequently used without their -ly endings: horrid embarrassed, for PROBLEM

SEMANTIC CHANQ£

191

example. Examine the following lisL of inlensifters and consider these questions. (1) Why have [here been so many V (2} Which arc stiil active? (3) What sort of semantic meaning did they have originally? (4) What has become of the semantic content of these words today? How do the example phrases prove Lhat in many casei; the words arc entirely lacking in any of their original semantic content? Following the example phrase is its earlier meaning. TABLE 7Л

INTENSIFIERS

Intensifkr

Example Phrase

Earlier Meaning

1, abominably 2. amazingly 3. astonishingly 4, awfully 5. confounded 6. damned 7. dreadfully 8. enormously 9. excessively 10. extraordinarily 11. extravagantly 12. extremely 13. fabulously 14,, fantastically 15. full lfi. frightfully 17, grandly 18, greatly 19. hideously 20. horribly 21. horridly 22, hugely 23, immensely 24. immoderately 25. incredibly 26, intensely 27. magnificently 28, marvelously 29. TnighL(il)y 30, monstrously 31. outrageously 32. perfectly 33. pretty

abominably tired amazingly dull astonishingly minor awfully unimpressive confounded happy damned pleased dreadfully small enormously good excessively happy extraordinarily normal extravagantly unambitious extremely limited fabulously cheap fantastically real full wearyfrightfully boring grandly expensive greatly exhausted hideously expensive horribly expensive horridly embarrassed hugely amused immensely unimportant immoderaLely hungry incredibly real intensely tired magnificently pleased marvelously stupid mighty weak monstrously excited outrageously ordinary perfectly stupid pretty ugly

(excite disgust) (cause to lose wits) (stun with a blow) (cause dread or awe) (bring Lo perdition) (doomed) (excite dread or awe) (abnormally evil) (go beyond just limits) (beyond the ordinary) (wander beyond bounds) (uttermost) (fabled) (exist in imagination) (complete) (full of horror) (great) (great) (excite terror) (excite horror) (excite horror) (immense) (large beyond measure) (beyond limits) (not to be believed) (stretched) (greatness of achievement) (miraculous) (power) (deviating from natural) (exceeding limits) (complete in all senses) (firm, nice, proper)

192

WORDS AND MEANINGS

TABLE 7,1 continued 34. powerful 35. really 36. right 37. simply 38. sore(ly) 39. stupendously 40. tremendously 41. terribly 42. terrifically 43. vastly 44. very 45. violently 46. wonderfully 47. wondrously

powerful tired really imaginative right foolish simply confusing sore(ly) afraid stupendously alert tremendously calm terribly pleased terrifically happy vastly insignificant very false violently opposed wonderfully boring wondrously tedious

(force, influence) (objectiveexistence) (what is good) (without complication) (cause pain) (struck senseless) (excite trembling) (excite terror) (excite terror) (of great dimensions) (true, real) (producing injury) (causing astonishment) (causing astonishment)

(And of course, there is the chief obscenity of English whose meaning has so faded that it could modify chaste.)

Minor Types of Semanlif There are some minor types of semantic change that are more curiosities than anything else. They have contributed a relative handful of new meanings to the language. Here are three of them. Folk Etymoiogy Often, a sneaker who misunderstands or fails to recognize a strange or foreign word will change it to one he dues recognize, introducing elements of meaning front the known word into the meaning of the original word'. Thus sambHnd (half-blind) became sandblmd. Shamefast, meaning fast in shame, became shamefaced: engn&g! (grig = tight, painful + n&gl = iron nail) became hangnail; berfrey (tower) became belfry; umbhpie (a pot pie made out of the umbles, i.e., the entrails of deer) becomes /tumble pie. PROBLEM 7.32: Here are some. Original words that have changed to a new form with a modified meaning resulting from the English associations of that new form. What are the new forms and how do they influence the meaning of the words? cucaracha, crevice (a kind of crustacean that looks like a tiny lobster), fcmelk (little woman), couteias (little knife), коаЫа ikooi = cabbage + sla = salad), brydguma (bryd = bride + guma — man), oichek (Algonquian for a little forest animal), surloln (sur French for above), pentfs (from Old French apentis, a small building dependent on another building, a lean-to), musquash (Aigonquian for a small water animal), chaise tongue (long chair),

CHANGE

193

techy (from Old French techs, spot or blemish, meanings which changed to peevish or vicious), wealhfmutu (wfalh = foreign + hnuiu - nut), Bursts (past participle of surseoir, to pause or leave off). (Tl has been claimed that woman is compounded from a phrase that meant woe + man, a folk etymology which reveals more about the etymologize г than the language. It derives from wif+ mann.) Semantic Replacement In our linguistic history, there have been cases in which the reverse of metaphor occurred. Instead of borrowing a foreign word for a. new semantic space, late ОЕ speakers transferred certain Danish meanings to Lnglish words which phonologically resembled the Danish words and shared some semantic space with them. The native English word dream, meaning to make a joyful noise, for example, was probably close enough to the Scandinavian word draumr, or sleep vision, to take on the Danish meaning, driving out the English meaning. The same may have happened with the words bread (originally meaning fragment); and dwell (OE meaning: to hinder or lead astray). Mistakes Another very minor source of new meanings attaching to old words is in the way meaning shifts from one to the other of a pair of words that are frequently confused. Thus What are you inferring? now frequently means What are you implying? Uninterested takes on the meaning of disinterested. Affect becomes effect. Others include continuous jron/inual, comparelcontrast, Some observers of the language expend great amounts of energy decrying such mistakes. It is not clear that their concern is commensurate with the- semantic confusion that may or may not result from the loss of a formal distinction.

Chapter 8 SEMANTIC CHANGE IN ITS SOCIAL CONTEXT

When we discussed the nature of borrowed words at the end of Chapter Five, we mentioned two kinds of forces at work in determining whether such words are retained in the language. The first kind of words are artifactual words, words to cover the categories that include objects and activities borrowed virtually intact from other cultures. Thus coconut, allegro, shah, and atall reflect little about our society beyond the fact that we are interested enough in Lhc referents of those words to need names fur them. The same can be said of certain kinds of word derivations: derrick, shrapnel,macadam, morel, radur. The second kind of words we called psycho-cultural, usually abstract words that reveal areas of meaning that penetrate to the psychological texture of our society: taboo, luck,ghetto*morale, chai№inismtbo«fdlenze,gestapo, and so on. Such words become an index to what is latent in our cultural psyches. We can distinguish some of these sane processes in semantic charge. Some meanings change because of artiFaclual reasons, because new objects come into being. Ship, sail, pile (as in atomic), broadcast, program, satellite, jet— all cover a wider category of technical referents, or artifactual referents, than they did a hundred years ago. We can discover a good deal about our technical progress if we examine the way such words have changed, but not very much, perhaps, about those unnamed semantic spaces that exist in our collective lives, №

196

WORDS AND WLANJNGS

VALUb JUDGMENTS Other kind? of changes do illuminate social and cultural aspects of our lives. The wholesale transfer of words in Western European languages that refer to physical actions to name cognitive processes {comprehend, explain, conclude, grasp, get, understand, and so on) suggests a good deal about the importance of self-conscious cognition in our culture. When words narrow to specify a particular kind of dog: bound] or animal: deer: or food: /neat; or dope: grass; and so on, it suggests something about what many speakers are preoccupied with. The transfer of sense words {sharp, sour, low, bright, for example) to personality suggests something about our need for a rich vocabulary lo discuss how we feel about one another The extensive vocabulary, largely metaphorical in origin, that describes sensations of taste, hearing, touch, and so on indicates how important that kind of experience is to us, A relatively significant increase in the metaphorical words for sounds in the last two centuries reflects a heightened interest in music. PROBLEM S i : Here are some words that cover semantic space in the field of female reference. Following each word is one or more earlier meanings. Comment. L shrew: a malicious, eviJ, cunning man 2* termagant; a male Saracen idol 3. harlot : a you n g, base fel low 4. hoyden: a boorish male peasant 5. scald: from Old Norse, a poet or lampooner 6. baggage: a wo rt h less fe I lo w 7. frump: a derisive snort > a jeer > 111 humor > a cross, dowdy man or woman 8. brothel: a worthless debased fellow > a prostitute > hence brothel house > brothel 9. bordello*, originally horde!, a villein of the lowest rank, hence horde! twuse, hence current meaning 10. bawd: a male or female procurer IL <:hit; а у oung boy о г g i ri 12. witch; originally either male or female 13. gossip : god s i b, or god-rehni ve > a familiar acquaintance 14. jade: л worn-oui horse 15. tart: short fо г sweetheart 16. virago; "The Woman" (Biblical) > manlike heroic woman > bold, impudent woman 17. quean (obsolete): woman > whore 18. iw лей: а у oung £i r L о г worn an

SEMANTIC CHANOE IN ITS SOCIAL CONTEXT

197

19. hussy: shortened Form 20. courtesan: a woman attached to a royal court. (This was the original Italian meaning. The MudF. meaning has been the dominant one in English,) 21. mistre$3'. feminine nf master 22. madam: ma (my) + dam (dame) 23. dame: from Latin domino, & woman of rank heading a household 24. #J>/ : a ch i 1 d of ei 1 her sex 25. boy: a fettered person > a servant (with vague contemptuous meaning) > lower class male 26. cuckold: allegedly from the cuckoo bird, в bird which lays its eggs, in another's nest. (BLJL the word refers to the husband of the unfaithful wife, not the ma!e who is the cuckotder. Note that we have no current word in the language for the betrayed female, though at one time we did have cuckquean to refer to the gulled wife,) Here are some other insulting terms for women: brand, chippy, drab, floozy, slur tern, slut, strumpet, trollop, trull, trou doxy, hag, harridan, crone, biddy, harpy, vamp, /tag, whore, bitch, piece, lay, tail, truff, hen, aid maid, wallfinwer, unladylike, unfemimne. (I) What aspects of feminine reference do these words focus on4.' (2) What are the masculine counterparts for these words? Are there as many insulting names for men as for women? (!) What kinds of semantic changes are at work here? (4) In what ways can lady be i as u I Ling (as opposed to woman) ? PROBI.FM 8.2' These words have been used lo express social judgments, usually of men. Earlier meanings follow each one. Comment. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

boor: peasant knore: a boy, servant churl: lowest rank of freeman ™r/f/; u n a ttend a nt or servant clown: a rural person rascal: rabble pariah: from the Tamil language in India, an outcast of the untouchable class idiot: ignorant blackguard: scullery assistant lewd: of the iaity, not a member of a religious order > ignorant common: a harcd by al I vulgar: ordinary language > persons belonging to the ordinary class villain: a low born person «Wv:Sire-ly mean: held by two or more (the sense intended here is of a mean hovel) coar.ie: common

198

WORDS AND MEANINGS

17. henchman: hengest (horse) + man {agroom) What kinds of semantic changes probably created these meanings? PROBLEM 8.3: What animal terms have been used to express value judgments'.' Ask someone who speaks another language to compare the derogatory and laudatory animal terms in his language and in English. Do other cultures choose animal terms for athletic teams 7 When the meaning of a word narrows toward an unfavorable meaning, it is called pejorative change. When the meaning narrows toward a more favorable meaning, it is called elevation or amelioration. We have already seen this happeit with words like merit, praise,fortune,fame, and so on, words that originally implied only neutral judgment, Lhcn gradually narrowed in the direction of [+favorable]. Л process similar to this has occurred with words that refer to various occupations and social roles that have elevated through history: knight originally named only a boy or youth; constable, the head groom; queen* a woman; minister, a servant; marshal4, a stable servant; lady, a kneader of bread (?); steward, possibly overseer of the sty, i.e., sty + ward; ambassador, a messenger: engineer, a plotter or schemer; baron, a feudal thane. Any of these could have changed in an unfavorable direction.

(189,226)

EUPHEMISM There is another kind of linguistic amelioration specifically directed toward finding socially acceptable words for concepts that many people cannot easily speak of. It is called euphemism, from the Greek for " t o speak favorably." It recalls an earlier attitude toward language —that somehow between 4he word and the thing there exists a bond so strong that (he word is virtually equivalent to the thing. So instead of calling cancer cancer, we call ha growth, almost in. fear that calling it cancer will make it so. A liquid which a woman sprays on her genitals in the belief that they will then not smell bad becomes a feminine hygiene spray. An invasion in Southeast Asia becomes an incursion; retreat is called mobile maneuvering, A person does not lie in his box where people can come and look at him before the undertaker puts him in a hole in a graveyard. The departed rests in his casket in a slumber room where friends may visit with him before the grief therapist assists the dear one to his plot in a memorial garden. Euphemism is such a pervasive human phenomenon, so deeply woven into virtually every known culture, that one is tempted to claim that every human has been pre-programmed to find ways to talk around tabooed subjects. One of the oldest known euphemisms is the word we use lor bear. It once meant

SEMAM1IC CHANGE IN ITS SOCIAL CONTENT

199

"the brown one," and goes far back into Indo-European prehistory when animals were probably totems and their names forbidden. The biological foundation is probably in the remaining connections between man's [imbic system and his speech system. (Review pages ]&-2GJ) PROBLEM 8.4: Here are some words selected by speakers of English as substitutes for more offensive or imiudicicnlly prestigious words. What areas of our culture are involved ? Try to discover the semantic and grammatical processes by which euphemisms are created. Ff the meaning is unclear, a definition follows. 1. 2. 3. 4. 5. 6, 7. 8. 910. 11. 12. 13. 14. 15. 16. 17. IS* 19. 20. 21. 22. 23. 24. 25. 26. 21. 28. 29. 30. 31. 32. 33, 34.

little girl's room puss on (die) social disease expectorate glow (perspire) growth (cancer) memorial pork (cemetery) human waste (feces, urine) exceptional child (m oron) sleep with grief therapist (undertaker) centr al at у (si u m s) go fa the bathroom In dy of the evening (prostitute) sanitary engineer (garbage man) senior citizen have ivgo (need to excrete) sexual relations rojrf (urinate) culturally deprived building engineer (j a ni tor) comfort station expire (die) capital punishment inter sen! (buttocks) washroom funeral home upchuck (vomiL) terminalcase blossatn (pimple) hfckey (pimple) с as unitу (к i 1 led i n actio n) garment (girdle)

35. 36. 37. 38, 39,

dental plates (false teeth) bridge (false teeth) bra panties monthly difficulties (menstrua-

40. 4), 41. 43. 44. 45. 46. 47. 48. 49. 50. 51, 52. 53. 54, 55. 56. 57. 5S. 59. 60. 61.

male organ (penis) Hafisen's disease (leprosy) hairpiece (wig) dentures heart condition (heart d i sease) tipi-y (drunk) disabled (crip pled) special student (retarded student) job action (strike) love child (bastard) plump know (have sexual rel a ti tin s wi th) pre-owned(second hand) sanitary napkin slumber robe (shroud) fix (castrate) swinger (promiscuous) irregularity (constipation) matterIpus) VD (gonorrhea/syphilis) В M (bo wei movemen l) final solution (extermination of Jews) criminal assault fanny (buttocks) wee wee (urinate) put to sleep (kiil) underprivileged

tion)

62. 63. 64. 65. 66,

WORDS AND MEANINGS

To create euphemisms like these, we can rely on any one of five processes. The first process i* widening. This blunts the impact of the semantic features being communicated in a single word by moving up one level of generality to name the superordinate set, usually omitting the specific feature that would unequivocally identify the referent. The specific cancer becomes the more general growth; the specific pus becomes the more general matter; the specific urinal ion becomes the more general voiding, the specific girdle becomes the more general undergarment. In some cases, the general features are split between two words. Hitting someone on ihc head is a criminal assault, but by convention, we know that this very general Icrin refers to rape. Many diseases are social diseases if we contract them through social contact. But syphilis and gonorrhea are particular kinds, and by convention we know social disease means VD, itself a euphemism for venereal disease, another euphemism, for the names of the diseases themselves. Expired breath, heat and sweat are all human waste products. But human waste usually means only urine and feces, themselves euphemisms for/;ш and shit. lit other cases, the twu or more words do specify the exact subset, but splitting ihe semantic features between them still softens the impact: hairpiece, inate organ, capitalpunishment, mental retardation. Schematically the difference between one word and a more general construction can be illustrated like this: solid Г + concrete" -animate

[+solid

+

human + waste = "+ concrete " " +concrete] + animate J +produced -r human J — wanted

feccs " +concrete + solid + human/ produced — wanted + organic

Should there be doubt that spreading semantic components across several words rather than delivering them in a single word lessens the impact, one need only compare the relative emotional effects offeces, solid human waste, and that material of a nan-fluid, non-gaseous nature which is the byproduct of metabolic and digestive processes in higher order primates. The second process owes much to our propensity for borrowing words from Greek and Lalin. There is a scientific or learned word for every bodily and social function we can think of. Even bad breath has been renamed halitosis. Expectorate, perspire, expire, abdomen, penis, defecate, urinate, eructate, crepitate, ragina --all carry strong clinical scientific, medical overtones that render them, if not completely inoffensive to many, at least usable when speakers cannot avoid talking about their referents, The third process is identical to semantic shift, A word naming a part of the larger complex in which the element named by the distasteful word operates shifts over to re-name that element. Thus we make love, от sleep with

SEMANTIC CHANUE IN ITS SOCIAL CONTEXT

291

or go to bed with instead of copulate or fornicate (which has a curious history). Such terms name what comes, une might presume, before, simultaneously with, or after the act in question. Л grief therapist may attend to the emotions of the bereaved, but his main j o b is to dispose of the corpse. Thus a name for an element within ihe larger process diverts attention from the central element in the complex of dealing with dead people. The memorial park to which he drives the body is '* ready" a grmeyerd, a place for graves. Bui the subsequent action of *" memorial'" or remembering gives its name to the location. Going to the bathroom precedes the act referred 10 while a comfort station is identified by the condition resulting from the act performed there. The fourth process, metaphorical transfer, may not be as common as it once was as a source for the must frequently used euphemisms. <J/O*V for sweat and blossom for pimple are clear examples, along with slumber robe and swinger. Belly button and break wind themselves once euphemisms, now need their own. Actually, a good many of our current vulgarisms and obscenities for which we now need euphemisms, may themselves have originally been euphemistic metaphors. Cock is a metaphorical transfer from the spigot on a barrel: prick is self-evidently derived from the action. Cunt may ultimately derive from a word related to Latin runnus, related to cuntits* a wedge, though the source of the -/ at the end of the word is a puzzle, A more likely source is a Low (no pun intended) German word, kont, the female pundcndurn. Pussy seems self-evident, but it may also be simply a shift following a transfer since pussy, a sixteenth-century term of endearment for a girl or lady, may not have acquired its vulgar sense until the seventeenth century. (Bunny and coney, another word for rabbit Eind pronounced cunny, went through the same series of changes.) The origin of twat is obscure, but it may be from iwachytfe, or a passage. Finally, fuck, that vulgarism of vulgarisms, is probably related to German fie ken, to strike or knock. It may have been borrowed from Scandinavian. Norwegian has & Jukka and Swedish zfocka, meaning to copulate. In any event, it definitely is not an acronym for For Unlawful Carnal Knowledge allegedly written over those put in stocks for that act in Puritan New England. With the permissiveness of the seventies, the rich vocabulary of expletives and swear words available to speakers at least from Chaucer's time has dwindled away until we are left with the tedious repetition of that one last Anglo-Saxon obscenity that can still infuriate most of middle America. Some have claimed that we need a new vocabulary, suggesting we try out words Jibe •xphugw
202

WORDS AND MEANINQS

crimanentHes, for crying out loud, son of a gun, gol dufned, gee whiz, for Pete's sake, for the lor? af Mike, jimminy cricket, jrtrpcrs creepers, dang, dud-bfmne, dad-burn, sont hit!, land's sake. The seventeenth century had a richer set than we because the < h real" name of God was banned from use in stage plays or In jest; so new ones were required: 'zounds (God's wounds), 'zbfood (God's blood), 'sfoot (God's foot), 'sbodies (God's body), "sdcath (God's death), 'shear!, bygog, by соске, bycrackey, and so nn. Other kinds of phonetic distortions draw on the diminutive -yjie ending: parity, tummy, hicky, fanny, or on the reduplication of high vowels: pee-pee, wee-wee, poo-poo, doo-doo. And finally, there are the acronyms: VD, ТВ,

HO, GD, SOB,BVD, BM, WC, SNAFU. Some Psjcholoyiiiil Causes of l:uphcrni$m The meaning of anything can be psychologically (not linguistically) defined as the sum of our responses to it, whether the " i t " is an object or an abstraction. Thus the m e t i n g of lion could be defined as our response lo one. We have already noted the problems with this kind of psychological explanation for a linguistic theory of meaning. But it may give us an insight into this particular problem of the constant turnover of euphemisms. The simple 5 -> R theory was modified to claim thai the responses lo a •word wai a "disposition" to respond rather than the response itself, a claim that was modified again: Rather than a disposition to respond, meaning became a mediated response. That is, afler a word has been associated with ihe stimulus lon
>•( Semantic Stimulus (word) }

I ( Primary Response i Mediated Response (meaning) a b c d e f g h i

Mediated Stimulus

Г Response to Meaning )

•^

SEMANTIC CHANCE IN ITS SOCIAL CONTEXT

203

For example, the action of vomiting elicits & primary response composed of elements u-i. The word ootnil becomes lirmly associated with the experience vomit, becoming a new stimulus that elicits some of the elements of the response that vomiting itself elicits; let us say ch e, f, and g. This meaning becomes, in turn, a stimulus which causes us to respond in certain strong ways not only to vomiting itselfbut lolhe word vomit. Ruphemisms arise when ihe nasty elements of a response attach so strongly to a particular word that the word elicits too many of the same responses as the thing. Puke and vomit elicit distinctly unpleasant responses. So we substitute a word or phrase that has not yet gathered to itself those associations: upchuck, sick to my Stomach, and so on, But eventually, through continued association, too many of the unpleasant responses to the referent will attach to these words and yet new euphemisms will have to be found. 8.5: What would be the nature of language if it were totally independent of our limbic system ? M 8.6: Assume that for same reason, eating and all its associated objects, actions, and ideas become a tabooed subject. Make up euphemisms, using all the processes discussed above to substitute for words like mouth, teeth Jerk,dinner, food, and that ultimate obscenity, eat.

SLANG, ARGOT, CANT, JARGON Just as some words are used by certain social classes Eo label other classes, so other groups of words are used by certain speakers to identify—though tacitly themselves. We have already seen how English lets us modulate tone through two or three styles by choosing different words: 1. 2. 3.

Professor Smith gives failing grades to all lackadaisical athletes, Smil h fl un к s e very la zy p Lay с г. Smit ty axes any j ock w ho goofs off.

From the point of view of someone in or out of college life, the difference between fail, /funk, and ax is more than just a difference in style. Flunk is more likely to be used by students than by teachers; and ax (orfox, meaning " F " ) has been, or was, used almost solely by some students among themselves. Other terms have also been used to identify words associated with particular groups. Along with slang, they simultaneously identify a vocabulary and implicitly judge it. Cant (related to chant, and originally the whining pleas of beggars) is often used to refer particularly to the language of thieves, gypsies,

204

WORDS AND MEANINGS

and such. But it has also been used to refer to the specialized language of any occupation, particularly to the mechanical and mindless repetition of special words and phrases. Argot {a French word of unknown etymology), usually refers to the secret language of the underworld, though it too has also been used to refer to any specialized occupational vocabulary—the argot of the racetrack, for example. Jargon (once meaning the warbling of birds) is usually used by someone unfamiliar with a particular technical language to characterise his annoyed and puzzled response to it. Thus one mar's technical vocabulary is another's jargon. Feature, shift, transfer, artifactuat, narrowing, acronym^ blend, dip, drift- -all these words belong to the vocabulary of semantic change and word formation, the vocabulary of historical linguistics. But for anyone ignorant of the subject and unfamiliar with the terms, such words would make up its jargon. Thus cant, argot, and jargon are words that categorize both by classing and by judging. Slung (of obscure origin) has many of the same associations. It has often been used as a word to condemn " b a d " words that might pollute "good" English—even destroy the mind. For Oliver Wendell Holmes, the use of slang was " a t once a sign and a cause of mental atrophy." For George Kittredge and James Gneenojgh, authors of a classic work or English words, (74) slang "has a taint of impropriety about it which makes it offensive . . . slang is illadapted to serve as a medium of intercourse and therefore is unsuitable for adoption into legitimate speech . . . ; the unchecked and habitual use of slang (even polite slang) is deleterious to the mind." But for others, slang is a technical term like the terms grammatical and ungrammatica!—a neutral term that categorizes a group of novel words and word meanings used in generally casual circumstances by a cohesive group, usually among its members, not necessarily to hide their meanings but to signal their group membership. It is in this second sense that we shall use slang here. Since. the same processes at work in "respectable" word formation and meaning change are also at work in slang, slang is probably as old as language and a source of meaning change. Some of the earliest recorded instances of slang go back to Aristophanes (c. 44S-C. 388 B.C.}. In his play The Frogs, a character is said to be "cracked," Certain words in Homer (c. ninth century B.C> have been identified as possibly slang. Among the classical Roman writers, slang can he found in Plautus, Horace, Juvenal, and Pctronius. It can be traced through Vulgar Latin to the Middle Ages. In France, the earliest records of slang date from about the middle or" the fifteenth century. In the late fifteenth century, a glossary of Rotwalsh, or German thieves slang appeared. In England, the earliest vocabularies of thieves' slang appeared in the sixteenth century: Copland's The Hye Way с to the Spyttel House: John Awdeley's Fraternitye of Vagabondcs, and Thomas Harman's Caveat for

SEMANTIC CHANCE IN ITS SOCIAL CONTEXT

295

Common Cursctoiirs. By the end of the sixteenth century, Robert Greene was recording thieves' argot in his Coney-Catching Pamphlets. And by the eighteenth century, Dr. Johnмп was noting in his dictionary his judgment of a word's social propriety. 8.7; Coliect twenty or so words that you would consider to be your contemporary slang. Why do you identify some words as slang; It's о groove (if that is still fashionable) and others as not; it*s a bore. Using the model of meaning change and word formation provided in the foregoing pages, explain the processes by which your slang came into being. Are there any slang words you can nor explain using that model? What areas of meaning are most frequently referred to? What areas of meaning have few or no slang words associated with them ? Why ? PROBLEM 8.8: Here are some American "slang" words that are iistcd. in Eric Partridge's Slang: Yesterday and Today (New York, 1970). Words that require explanation have a meaning in parentheses. What sorts ofqucstions does this list raise? 1. 2. 3. 4,

5. б. 7.

8. 9.

10. 11. 12.

13. 14. 15,

16. 17. 18. 19.

20. 21. 22.

23. 24.

25. 2*.

artillery (beans) bangitp (exactly right) chuck (food) darb (a popular person) fade (a poorly dressed person) #(>o/(sweetheart) ' hipe (cheat) ivories (teeth) charge (injection of narcotics) kale (money) lilies (hands) mess (dull person) nines (the limit) peav.h (pretty girl) all in (tired) alt set (ready) 6ee/(complain) big shot baloney (nonsense) brainstorm broke (out of money) buddy butt in chow crap crasher

27. divvy up 2Я, dough 29. dude 30. flirt 31. fuzz 32. dope (foolish person) 33. straight 34, bleachers (uncovered grandstands) 35. bluff 36. blurb 37. booster 38. boss 39. camouflage 40. catch on 4 1 . close up (as in movie making) 42. fade out (as in movies) 43. stilt (и$ in movies) 44, fake 45. insider 46. frolic 47. gadget 48. hike (walk) 49. hoodlum 50. leak (tell a secret) 51. phony

206

WORDS AND MEANINGS

52- prom 53- proposition (matter to be considered)

54. prowler 55. turndown

PROBLiiM 8.9; What ultimately happens to slang words? Consider the foregoing list and Johnson's list of "low" words in Problem 7.9, Es there any other list or lists of words provided so far that might conceivably be relevant to this problem of slang? Given the short life of any given set of slang words and their appropriateness to tather narrowly defined social situations and social groups, it is not surprising that slang should be condemned by mosL teachers. For they are almost exclusively concerned with teaching a kind of standard language appropriate to communicating with an unknown audience that may be reading a piece of ptose long after it was written. Despite the lively, vivid quality of mnch slang, if it is too evanescent to be undersLood after a relatively short time, к fails to communicate clearly. And even if the reader understands the words, nothing bugs a dude like a cat who lays slang on you after it is out of date. Whether slang, like syphilis or heroin, atrophies the mind is, of course, another question. Slang not only satisfies a quite legitimate desire LO play with language (why else puns, tongue-twisters, palindromes and so on?), it also often provides a word or phrase for a category ofmeaning that ''respectable" word*s fail In cover, ft is true that mug, lilies, dogs, schnazzala, peepers, gams, balls, yap, ivories—all current or dated slang words for parts of the body— express no new referential area. But they do invest the category of items with a particular affective tone. If we lost these words, we would still have face, hands, feet, лаге, tyes, legs, testicles, mouth, teeth, and so on, But without slang terms for them, we would be unable to express the affective meaning we invest in them. But some words name a category of experience that no standard word covers: graft}fan, doodle, leak, turn off, dig, hip, frisk. In such cases, if the word really does meet a felt or even tin felt need that is not just a temporary part of the cultural vocabulary of a limited group, then the word stands a good chance of being absorbed into the more permanent vocabulary of the language. In fact, when we consider the original mcLaphorical transfers of words like resuli, explain, comprehend, grasp, and so on, we might suspect that even they were not as formal and academic when they were first used as their literal use is for us now. They have become necessary words because the semantic space they now occupy is important to a society preoccupied with thinking. And because thinking is important, the words have become respectable. Thus what may begin as slang may take one of four courses. First, and most likely, it may disappear entirely: skidoo is kept alive only as a trite

SEMANTIC CHANGE ПМ ITS SOCIAL CONTEXT

207

example of archaic slang: gvvf for sweetheart, grummy for low spirits, rake for comb have almost certainly disappeared everywhere. Second, a slang word may move inLo a limbo between slang and very casual speech: booze, dude, mob, junk, chintzy, fire (dismiss from a job), dumb (stupid). Third, other words that once may have been slang-like may move more into the permanent casual vocabulary of the language: cab, taxi, bus. Night, nag, fun, wail, boater, bigot, flimsy, flippant, sham, shuffle. Even the most fastidious speaker would be unlikely to boggle at using such words in casual moments. And finally, there arc those words that have been completely absorbed into the language, into an English appropriate to writing, of this sort. Oniy the most super-sensitive soul would object to words like these: A sensible student does, not gamble on a future job by dressing in a shabby manner. Nor does he appear touchy when asked to volunteerfor extra work. The words in boldface were listed by Dr. Johnson as " l o w " or''ludicrous" o r < l c a n t " words. Other words he condemned included bolster, abominable; desperately, finesse, fragmentary, slim. Others have condemned banter, fop, flippant, bigot, and flimsy. Yet today we would not be shocked to find all these words in quite elevated prose. (52,75,159,162)

SEMANTIC LAWS Historical semanticists have searched for laws, or rules of semantic change, much as historical linguists studying sound patterns and grammatical structures have. In the nineteenth century, a group of Germanic philologists were able to formulate iaws of phonological change in the Germanic languages that were of very great explanatory power, laws we shall explore in Chapter Thirteen, Inspired by their success, semanticists set about looking for equally powerful law* of semantic change. Unfortunately, they have met with very limited success. There are a few tendencies that we can discover in the previous data: 1, 2, 3, 4, 5,

Words for abstractions will generally develop out of words for physical experience: comprehend, grasp, explain, and so on. Words originally indicating neutral condition lend to polarize: doom, fame, predicament, luck, merit, Words originally indicating strong emotional response tend to weaken as they are used to exaggerate: awful, terrific, tremendous. Insulting words tend to come from names of animals or lower classes: rat, dog, villain, cad. Metaphors will be drawn from those aspects of experience most relevant to us: eye of a needle,finger of land; or most intense in our experience: turn on, spaced out, freaked out, for example.

20Н

WORDS AMD MEANINGS

But these only approximate statistical tendencies. They are not laws fulfilling the usual requirements of a law: that it not only explain relevant past events, but predict future events as well. The odds are thai a word for an abstract cognitive or emotional process will develop out of an earlier concrete reference, but we cannot take any given word with concrete reference: rock or push and predict that it will evolve into a word referring to abstract experience. And the reverse has happened with originally abstract words like did], blunt, keen, and soft. All originally referred to nonphysical referents. A change that has been given the name of " l a w " and often cited as one of the best examples of a law is thai, suggested by Gustaf Stern: Adverbs meaning rapidly before 1300 developed into adverbs meaning immediately. Adverbs meaning quickly after 1300 did not. This, of course, is peculiar to linglish and limited to a particular point in time. At best, it is a restricted generalization about the past, not a law than encompasses the future. (2(Ж) We m и si, however, distinguish two kinds of predictions: (1) Tn some cases, we have to base a prediction on relevant external forces. A culture will invariably assign meanings it must develop to the words it uses most frequently or can create most easily. Thus a generalization about, say, euphemism,, is enLirely cultjre-dependent in regard to the tabooed and non-tabooed areas in which euphemisms will be used, (2) On the other hand, there may be universalsof change which, because they are so regular and so general, reflect internal influences either peculiar to the particular language or Lo human language and cognition in general. It has been claimed, for example, that all human languages obey a very general law for the development of color terms, (9) The development has seven stages and each stage must be passed through in sequential order. The order can be represented like this; purple " white'

' green

' yellow "

pink J ^ blue > brown >

_ black _

yellow

green

orange _grcy

I

II

111

IV

V

VI

VII

That is, all languages have nonmetaphorical words for black and white. Such words may exhaust the spectrum. If there are only three terms, the third one will be a word for something approximating red. Stages III and IV may occur in either order: green and yellow will be the next words to develop, but they may develop in either order. Once those five words have developed, then blue and brown will follow, themselves to be followed by the next four in no particular order. As the authors of this research have suggested, there arc eleven universal

ЗВМАНПС СЙАМОЕ 1И ГМ SOCIAL CONTEXT

209

potential color categories, and those colors develop in a fixed order. If they are correct, this is surely an example of cognitive-linguistic interaction and a very powerful semantic " l a w " dependent not on any particular cultural organisation but rather only a particular level of cultural complexity. There may be another law of semantic change in English where mental processes interact with a culLural milieu to create new meanings for old forms. Words fur sensory experience fall into five basic categories. Tactile: mild, soft, smooth, even, hard, rough, harsh, course, dull, dry, keen, sharp, hot, cold, warm, cool, heavy, light, dry, bland, hard. Gustatory /Olfactory: piquant, pungent, tart, bitter, cloying, ееfid, acid, sweet, sour, vapid, tangy. Visual (dimension): high, low, thick, thin, deep, shallow, wide, broad, narrow, full, big, little, fluf, steep, small, level. Visual (color): clear, light, bright, brilliant, fair, dark, dim, faint, pale. Aural: quiet, loud, shrill, strident We can expand these basic categories by transferring words from one category to another: hot music, laud colors, sharp tastes, sour smells, and so on. But not all metaphors are equally comfortable: loud heights, wide smells, bright edges, low tastes (i.e., parallel to low sounds on a sensory dimension not the sense of vulgar tastes). At lirst glance, we might assume that we randomly seiecE our metaphors, since many of these are perfectly clear even though we never or rarely use them. A loud height, is certainty higher than a quiet one. But the historical development of these metaphors in English reveals a pattern that approaches the regularity of a law that might be represented like this:

COLOR

UCH

> TASTE

--> SMELL

DIMENSION

\

SOUND

210

WORDS AND MEANINGS

That is, in English (and possibly many other languages), words referring primarily to tactile sensations will, if they transfer at all, transfer lo tasLe

(sharp taiies), color (harsh rotors), and sound {soft sounds). Taste words will transfer to smell {soursmell) or to sound (sweet sounds). There are no primary, non-metaphorical smell words in English (or in many other languages, interestingly enough). Dimension words may transfer to color {deep red) or to sound {high sound). Color and sound words trade metaphors (dark sounds, quiet colors). There is a second regularity: Each successive transfer follows Lhe same sequence. That is, ones dull transferred from tactile to vision, it transferred on to sound, but not back to taste/smell or dimension. Each successive transfer of a word is determined by the most "advanced " meaning of the metaphor. And there is a third regularity: If a sense transfer violates either of the first two regularities, that sense development tends lo disappear, hyr example soft taste, a sense not usual in standard English, was active for a time after soft sounds had developed, a violation of the "most advanced sense'1 constraint. That taste meaning subsequently disappeared. Shrill developed touch and taste senses, contrary to the main principle. Neither sense has maintained itself in the language. There are a very few exeepLions to this generalization: Mellow developed senses out of the predicted order; faint developed a smell/taste sense after earlier transferring from the meaning "feigned" or "simulated" to color. Sharp transferred lo angles, perhaps on the model of Latin acute. Flat and thin transferred to taste. But when all the transfers are taken into consideration, these exceptions constitute less than one percent of the total data. This kind of research into the semantic development of an entire field of words as opposed to tracing the history of a single word is little explored at the moment, largely because the data is so difficult to assemble. We require •what we do not yet have for any language: An historically organized Ruget's Thesaurus a resource that would list every word that has ever referred to any sense listed under the category of that sense along wiLh the earliest and latest citations of that word wiLh that sense. Without such a resource, historical scmanticists are forced to read dictionaries entry by entry, searching for senses and words that at one time may have referred to a particular referent but no longer do. How, for example, would we discover all the words that at any time have ever referred to, say, taste? These have: aspre, coarse, cold,, poignant, rough, smart, soft, warm, acrid, austere, dulcet, eager, acute, high, small, loud, and shrill. 8JO: Here are some sense words with their etymological sources. Docs this information confirm or contradict the generalization about sense transfer? iweet (pleasing to any sense), bitter (from biting), ocid (from ac- = sharp), acrid (from ac- = sharp), tart (sharp pain), cloy (prick with a nail), PROBLEM

SEMANTIC CHANGE IN ITS SOCIAL CONTEXT

211

piquant (piercing or stinging), pungent (to prick), stir it! (rasping or grating, earlier to scrape or abrade), quiet (calm, peaceful), dulcet (sweet to taste or smell), blunt (unclear of sight), keen (intelligent), soft (producing agreeable sensaLions), coarse (ordinary), dull (unintelligent), harsh (rough and hard to 1 touch, then to taste). How do these fit in: eager, poignant, austere, brisk } PROBLEM 8.11: Many of these words also apply to abstract situations and to personality: о dark moment., a rough time, a warm personality;, a dry humor\ and so on. Are there any general]zations to he made regarding the sequence of development of these transfers? You will have to consult the OED for dates-

PART III

GRAMMAR AND SOUND

r1

Chapter 9 BETWEEN SEMANTICS AND GRAMMAR

Between questions that are purely semantic—the origin of insulting names for women, for example—and purely grammatical questions, whether -tiort or -ness goes with happy—is a kind of language change that seems both semantic and grammatical. Starve, for example, originally meant simply to die. It later narrowed to mean lo die from hunger, hul it changed in another way. In its original sense, it was always intransitive. That is, it was possible to say He starred, but according to the O t D not until the sixteenth century do we find starve used in its transitive sense: Someone started him. Before that time, presumably, it would have been ungrammatical to utter that sentence, just as it is now ungrammatical for a ModE speaker to say, He laughed him, meaning he caused him to laugh. In order to understand this kind of change— and to understand even more complex problems of grammatical change—we have to pause for a moment to explore some of the ways in which such grammatical-scmamic problems are dealt with in a modern grammar, and how their solutions bear on problems of historical linguistics. GRAMMATICAL STRUCTURE Verbs are conventionally grouped into intransitive or transitive, depending on .whether they take an object: 215

216

1. 2.

GRAMMAR AND SOUND

The man disappeared, The manfixedthe роп о.

But most verbs in English are neither strictly transitive nor intransitive: За. 3b. 4a. 4b. 5a. 5b.

The people ate. The people atefooA. The man wa tched. The man watched the parade. The birdsang. The bird sang soni^s.

Dictionaries usually list these transitive and intransitive senses separately, as different meanings for the same word. Yet the differences may not be in the meaning of the word but in whether the word occurs before an object, before a noun phrase. PRORI.FM 9,1: Suppose we were concerned only with surface grammatical structures, with what we can observe about the order of words in sentences as they are actually spoken or written, with no concern for the meaning of a sentence. Would that attitude toward these next sentences lead us to some misleading conclusions about the notions transitive and intransitive? 6a. 6b. 6c. 7su 7b. 7c 8a. 9b. 8c. 9л. 9b. 9c.

Tom pours wifte. Tom pours. Winepoun. George eooks eggs. George cooks. Eggs cook, Bill drives cars. Bill drives. The car drives easily. Jack reads books, Jack reads. The book reads well.

If we were concerned only with surface patterns, we would conclude that the (a) examples were transitive verbs, verbs with objects: (b) and (c), intransitive verbs grammatically identical with the intransitive verbs in (10)-<J3): 10. Heexisls. 11. Tom grew. 12. Jack wailed. 13. The star twinkled.

BETWEEN SEMANTICS AND GRAMMAR

217

But the intransitive verbs in the (h) sentences are very different from those in (c), and both differ from (IO)-(I3): Those in (b) are implicitly transitive because in each case, we could put something after the verb, making it transitive without really changing the meaning: Tom smokes (something). But we cannot say * Eggs cook (some thing) or* He waited (something).*

And the (c) sentences differ from both (b) and (IO)-(J3): In each case, the subjects, wine, eggs, car, and bonk, are in some sense like the objects of the action. That is, they are being caused to do or undergo the action indicated by the verb: Xcauses eggs to boil—Eggs hail. The subjects in (b) and (10)-(l 3), on Lhe other hand, are not being acted on. So what appear to be grammatically similar sentences are really quite different in ways that we have to explain in grammatical terms, not just semantic terms. We thus have three kinds of transitive and intransitive verbs: 1. 1. 3, A, 5. 6,

Pu re i ntran s i ti vc (no objee L possi bte): George exists. Intransitive with implied object: The man eats. Intransitive with subject as implied object: The eggs cook. Pu re tran si ti ve {a hjee t general I y nequi red): Georgefixesthings, Transitive wit h о pt io na I object (see (2) a bove): 7йе тдл eatt 5<™ег№£Trans it i ve re Lated to (3) a bo ve: Someone cooks the eggs.

One new and very influential school of linguists has tried to account for these questions by assuming two levels of grammatical structure. One is the obvious sequence of overt elements we respond to in a sentence. The other level of structure represents the deeper grammatical and semantic relations that lie behind the surface of a sentence, relations we can describe independently of the overt, perceptible word order. The overt level we will call surface structure; the deeper level we will call the deep structure of a sentence. We can represent or li generate" one kind of simple deep structure behind sentences with formula-like rules something like the fallowing: 1. 2.

S > NP YP S = sentience, NP = noun phrase, VP = verb phrase (fN¥)\ V = verb VP ->• V IJ AP > I AP = adjective phrase, PP = prepositional phrase

3. 4.

NP -+ ART N PP - ^ p NP

ART = article, N = noun P = preposition

Rule (f) says that the abstract idea of a sentence is at the simplest level of organization made up of a subject noun phrase and a predicate verb phrase: Tom (eft, The. boy saw the dog, A girl seemed happy, The man mis in the house. \ In ihe nest three chapters * will mean nol a reconstructed ruol

218

URAMMAR AND SOUND

Rule (2) says that the predicate verb phrase can be made up of just the verb: left, in Tarn left. Or it can be made up of the verb and any one of the three elements in parentheses. Parentheses mean that whatever they embrace is optional. They may or inay noL be selected. The three elements are (in boldface) a noun phrase: The boy saw the dog: an adjective phrase: Mary seemed hungry; or a prepositional phrase: The man was in the house. The bracts enclosing the elements in rule (2) mean that one and only one of these elements may be selected if any arc selected. Rule (3) indicates that the noun phrase is made up of an article and a noun: tkeboy, a man. ait apple. Rule (4) indicates that a prepositional phrase is made up of a preposition and a noun phrase: in the house, by the road. These choices and the resulting structure can be represented on "trees." If only V is chosen in rule (2), then the tree would look like (a). If the NP were chosen in rule (2) along with the V, then the tree would resemble (b):

(a)

(b)

PROBLEM 9.2: What would the trees look like if ЛР had been chosen for one and PP for another ? Of course, these are only grammatical skeletons. We need semantic units, words, to fit under the itodes that specify parts of speech, under noun, article, verb, preposition, and adjective. These we can list in a dictionary that accompanies the base rules. In addition £o the meaning of the words (indicated in the form of semantic components) we list restrictions which determine whether the verb must or may occur before an object -.fixed the chair, before an adjective phrase: seemedШ; before a prepositional phrase: was in the house; or before none of these: He disappeared'. The dictionary component (minus the semantic features) might look like this. The position, specified by part of speech: disappear: V; become: V;

, indicates the position of the verb relevant to any other

f fix: NP/AP

V; be: V;..

NP _ NP/AP/PP

(The slashes without parentheses indicate an obligatory choice of one element.)

BETWEEN SEMANTICS AND GRAMMAR

219

That is, disappear occurs under a V but because it is purely intransitive, it may occur only without a complement. This restriction is symbolized by #. (This ignores modifiers of Lime and plate, and so on. which we shall omit for now): lie disappeared f. Find, a transitive verb, must have an object KP after it: He found money. Become, a copula verh, may have either an NP or an AP after it: He became kingltired. And be, another copula, may have after it any of the three: He isakingjddjm the horns. (31,94, 132,143) PRCMJL.FM У.З: Here are several words to add to the dictionary. Specify very simple restrictions for the verbs. Generate four or five deep structures, picking words at random. Where musL the dictionary information be modified, added to, expanded ? How might we do it ? N: man, cars, piano, food, parade, songs, wine, people, bird, Tom, George, BUS, Jack, eggs, bpoksrstart he, V: was, pours, drives, reads, exisrs, grew, н-aited, twinkledt became, cooks, exists* disappeared, fixed, ate, welched, sang, AP: old, heavy, smart, straight, hard, square, good. P: in, on, at,by. ART: the. PROBLEM 9.4: Add a rule that would account for these patterns: old, very old; If red, somewhat tired; nick, quite sick; intelligent, rather intelligent. Now the rules and the dictionary would let us generate the sentences mentioned above, (I) to (13). But if we just list all the different verbs and their possibilities under different entries in the dictionary, we will not capture a generalization about the differences and similarities among them, that He reads and He reads (something) arc somehow related, that Someone cooks eggs and Eggs cook are related. So even if any intransitive sentences such as He reads, Eggs cook, and They exist have the same surface structure, we also know that they differ from one another in very basic ways. Transformational generative grammarians try to account for all this by postulating transformations that change the deep structures generated by the kinds of rules described before Problem 9.2 into surface structures, into the firamnialical strings we understand in an actual sentence. The problem is to decide how surface and deep structure relate to one another. Consider Turn hunts, for example. It always implies Tom hunts something. We can capture that insight by assuming that the structure of Tom hunts is transformationally reiated to the structure of Tom hunts something, through a transformation that deletes the indefinite object in the deep structure. This produces a surface structure superficially like that of an intransitive sentence.

22(i

ЯЙАММЛН ЛМ) SE

We can represent the transformation rule like this: OBJECT DELETION:

V

NP (if NP is indefinite)

The iirst line refers to the relevant elements in a tree, the verb and [he object, and by implication, anything under those nodes. The second line represents the change, the elements in the new tree treated by the transformation. In this case, the whole NP node has been deleted. But if we applied this rule to all verbs that had an NP after them, we would delete any indefinite object after the verb. This would generate sentences like *Totn fixed, *Bi!lfound, * George earned. So we have to indicate that some verbs cannot have objects delcLed. Since most verbs do allow their objects to be deleted, It is simpler to stipulate the exceptions, those whieh тцу not u n de rgo о bject d e I et ion. We can do this by adding to the dictionary a fourth item of information. We ill ready include (I) part of speech, (2) semantic delinition, (3) the syntactic environment in which the verb may occur. We add (4), the transformations a verb may or may not allow. Thus for find, ™/, and exist, we would have dictionary entries roughly like this (semantic information is omitted): eat:

V;

NP.

exist:

V,

find: V: NP; -OBJECT DELETE.

§.

That is, find may not have its object deleted, while eat may; *Ttm found, Tom ate, Tomfound something, Tom ate something. Thus a complete grammar would look something like this:

BASE RULES DEEPSTRUCTURLS

TRANSFORMATIONS

DICTIONARY SURFACE STRUCTURES PROBLEM 9.5: One kind of transformation is DELETION: They eat something -* Tiiey eat. There are other kinds of transformations. Here are some other pairs of sentences that might be transformationally related. How? Which sentence Form is derived from which? Note: Do not assume that the first sentence structure is the source of the second. (Strictly speaking, one sentence is not transformed into another. The abstract structures behind sentences are transformed. We will use sentences here as a shorthiind for that structure.) 1. 2,

He looked the number tip. fie iookedup the number. Iknow that he is there. } know lie is there.

B6TWEFN SEMANTICS AND GRAMMAR

3. 4. 5. 6. 7. 8. 9. 10-

221

Л is eas у to do thai. To do thai is easy. Far someone lo do thai is easy. The man in the house left. The man who was in the house left. He usually leaves. Usually, he leaves, Down the street camea truck. A truck camedown the street. Alt of the people werethen. ASthe people were then. John and Mary married. John ant! Mary ntarriedeach other. For someone to do that is bod.To do that is bad. Tom is older than Jack. Tom is older than Jack is.

HISTORICAL IMPLICATIONS

The historical importance of these patterns when we describe them by this kind of grammar is this: Many individual verbs move from one class to another, thereby allowing a previously prohibited transformation. For example, as purely transitive verbs have been born or borrowed into tnglish, they have generally lost restrictions against deleting objects. The change in the grammar would mean that we simply remove the restrictions against object deletions. Verbs like paint, hunt, cook, read, eat, kill once obligatorily transitive -may now have their objects tick Led. But here we have to distinguish between the сопкещк-псе? of grammatical change and its causes. The consequence of the ehange is a new pattern for certain verbs. The cause of the change lies in us. Strictly tninsitive verbs regularly shift to optionally transitive verbs probably because we operate on a combined principle of economy and generality. ТГа verb requires an object, we cannot speak of doing whatever the verb indicates unless we specify a particular action or add an indefinite object. In ModE, for example, most of us do not ordinarily say *He is fixing around the house, but we may say He is painting around the house. Given the principle that we alt unconsciously strive for grammatical economy where it does not interfere with communication, we can understand why verbs referring to the most common and hence important actions are optionally transitive. Were they not, we could not in the most economical way speak in general about eating, drinking, reading, writing, and so an. On the other hand, verbs which still do not allow their objects to be deleted are 9 often too general to communicate anything without an object: He finds! detertvmes!prejcrx!mer:1ionslkeeps, and so on. Since only exceptional verbs arc marked against OBJECT DELETION, we might also consider whether another aspect of the principle of least effort, the simplicity principle, operates here. If the few verbs which may not lose their objects have to be specially marked in our internalized grammar, then a

222

GRAMMAR

child, as he learns verbs (or most adults as they use them, perhaps), may learn to use them without their accidental and often unpredictable restrictions, much as irregular verbs like help-halp-kolpen lost their exceptional status and regularized to heip-helptd-helped But simple object deletion does not explain patterns like those in sentences (€c)-(9c) tn a sentence lite The eggs cook, the apparent subject is a noun phrase that can be interpreted as the understood "receiver" of an action or as a deep-lying subject that is caused by some unstated agent to undergo or perform the action of the verb. If this is so, then we have to assume a different deep structure behind George cooks on the one hand and eggs cook on the other. We can get at the differences, perhaps, through history, PROBLEM 9.b\ Actually, four kinds of changes have created new transitive and intransitive verbs. The deletion of the object, as we have seen, is one. Compare these next groups of sentences. At an earlier time, the verbs in (I) and (2) were strictly intransitive; those in (3) strictly transitive. What wottid you speculate were the intransitive sentences historically related to those in (I) and (2); what would have been the transitive sentences related to those in (3)? Why did they change?

L 2. 3.

He whispered the words. They sang the lyrics. He answered his critics. He argued the case. He grows corn. He starved millions. We work our subnets hard. She dined severalfriends. The situation developed. The material disintegrated. The news continued. The troubles resumed. The waters separated. The oil spread.

Those sentences in (1) were originally pure in transit] ves, but eventually attracted an object, probably because the acLion referred to always involved some concept besides the subject, on the analogy of say something. Modern pure intransitives such as sparkle, kneel, disappear, exist, and die, on the other hand, ordinarily involve only a single noun phrase element, which is the subject. Those sentences in groups (2) and (3) historically constitute mirror images of each other. The verbs !n (2): grow, starve, work, and dine, were originally intransitives: Tomatoes grow, people starve, employees work, friends dine. Conversely, those in (3) have been cited as transitive before they were cited as intransitive: Someone developed the situation / disintegrated the material I continued the news / spread!lie oil, and so on. We have to explain the changes in two different but related ways. There is a large class of verbs called causative. Their subjects cause the objects to perform the action indicated by the verb. Compare the sentences in (a) with those in (b):

9

BETWEEN SEMANTICS AND GRAMMAR

a. b.

223

He rolled the bait (that is, he caused the hall to roll) / moved the box } broke the dish Ifried the eggs / rocked the cradle, and so on. He saw the ball I bought the box f found the dish I felt the eggs j painted the cradle, and sflor.

The subjects in (b) do not cause the objects to move or change, as they do in (a). In a generative theory of language, each verb of the roll, cook, move, break kind would have associated with it a feature we might call [causative]. Ordinarily, such a verb allows a transformation in which the underlying object is shifted to the subject position in the surface structure and the original indefinite subject is deleted: someone/thing ->• the dish

breaks the dish breaks

More recently, this has been modified to change the direction of the transformation. someone/thing (causes) [the dish (to) break] -> someone/thing breaks the dish The verb written here as (cause) is in this theory an abstract entity in the jdeep stracLunc that is not a real "word," but rather an abstract "operator" or theoretical element. nuL unlike certain related elements in algebra and calculus. There are a good many arguments in favor of this analysis, but for historical purposes, the simpler transformation will better account for the change. For we would otherwise have to assume an underlying intransitive break before it actually was attested to in the history of the language. Some transitive verbs like commence, circulate, increase, dissolve, continue, resume, reconcile, and so on originally had a causative sense in that they could occur in sentences like We continued the trial. But these causative transitive sentences did not at first allow an equivalent intransitive sentence: The trial continued. Before the eighteenth century, for example, it woufd have been as ungrammalieal to write The situation improved as it is now ungrammatical to write *The situation exacerbated. At the time, improve had to exclude the transformation which would have shifted the object, just as exacerbate does now. But apparently, just as speakers simplified their grammar by regularizing the OBJECT DELETION restriction, so they frequently regularized the restrictions in regard to CAUSATIVL OaJLCT SHIFTS by dropping the restrictions against it from verbs like continue and improve. The shift from non-causative inlransilives like those in (2), flowers grow, to causative transitives, I growflowers, requires a shghtiy dilic rent explanation, akin, perhaps, to back-formation. Intransitive verbs like grow, stanet work,

224

СЕЛМ.ЧДК AND SOiJNn

and dine, could occur in a pattern like The man starved or in literal causative constructions lite Someone caused the man to starve. But originally, they did not allow an implicit causative construction like Someone starved the man. Presumably, however, speakers added [causative] to the verb rather than keep it in the sub-class of verbs which did not allow a transformation of the types Someone Causes the platform to turn -* Someone turns the platform -> Tte platform turns. Thus when Flowers grow lent itself to a pattern like Someone causes flowers to grow (or the approximate idea), it could easily lead to Someone growsflowers on [he analogy of the verbs that do allow the transformation. The modern consequence of this is two identical pairs of sentences: Hegrowsj improves sppks and Apples grow Improve. But each sentence is historically the mirror image of tlie other member of the pair, So with two transformations: OBJECT DELETION and CAUSATIVE OBJECT SHIFT, we can account for a variety аГдавйгаротагу grammatical patterns. And in the terms of our theory, we can account for the historical changes in words that occur in these patterns: Speakers tend to simplify a grammar by removing unpredictable restrictions against transformations. In a sense, this is the grammatical explanation for analogy. PROBLEM 9,7: What do these sentences mean? 1. 2. 3. 4. 5.

* Heexistedthe newelement. * The child cried the woman. *She waited the people, * I disappeared she cat. * They t winkled the Ugh t.

Find five or ten other ModE verbs of the same type. PROBLEM 9.8: Some grammatical structures are semantically causative but their verbs do not allow the shift. For example, I caused the man lo leave does not allow / left the mart, as / caused the man to trip allows / tripped the man. Here are some other examples: / caused the man to losejsucceedlselljstudyj fightleatSspltjck'snldustlwisshlconfessjapplaudjexplaln. What might be one explanation of why these do not allow the transformation? Is this problem related to the immediately preceding one? What is the meaning of/failedmy friend'! PROBLEM 9,9: Does any of this explain sentences like these: This book reads easily: the car drives awkwardly; the class teaches welt; the turkey canes cleanly; the wood saws hardl Check the dates for these senses relative to others connected with same words in the OED. Comment. A stylistic consequence of these changes, perhaps even a secondary cause, is a greatly increased flexibility in how we can construct sentences in English

BETWEEN SEMANTICS AND GRAMMAR

225

to reflect rhetorical nuances. When verbs like those we have described occur in new patterns, in the language, they allow a writer to control rather carefully the rhetorical structure of his paragraphs. One organizing principle of discourse is the way a writer controls what words fill the subject position of a sequence of sentences within a paragraph. How well a writer can control his reader's attention by manipulating subjects partly determines the quality of his prose. This discussion should also point »ut how different theories of grammar and meaning will lead us to different histories of the language. If we follow a theory of traditional dictionary definitions of verb meaning or the theory of those who primarily collect grammatical patterns, the kind of changes that we have described here would be represented in a dictionary something like this: Earlier period:

teach;

verb transitive: ] teach English. verb intransitive: I teach.

Later peri od:

teach:

verb transitive: I teach English. verb intransitivej: I teach. verb imransitive2: The children teach ea.tily.

The dictionary simply adds a meaning to the word teach as its syntactic environment changes, A transformational generative grammar would argue that in these sentences the meaning of teach does not essentially change. Ail that change are the possible surface relationships among the noun phrases that occur with the verb in the deep slructunc. And these potential new relationships can be represented by minor notational changes in the dictionary. Here are four examples «f notational changes. (We omit everything except the relevant information. Jn each pair, (a) is the earlier entry, (b) the later.) explain:

a. V; b. V;.

siucty, a. V:. b, V;. improve:

a. V;. b. V;.

. NP; -OBJECT DF.LF.TION. (OBJ. DEIFTE now . NP, possible) •#•

. NP.

(direct objecL now assumed)

. NP; - CAUSATIVE SHIFT, (CAUS. SHIFT now NP. possible)

226

URAMMAH AND SOUND

grow:

a. V; b. V;

NP.

(CAUS. SHIFT now possible)

fin addition, we would add [causative] togrew.) Another kind of linguistic change which falls between semantics and syntax is illustrated by the difference between these sentences: a. It is made of stone. He eatsfruit. They / « song.

b. It is made of stones. He eats fruits. They lore songs.

In the (a) group, the noun b a пяп-counf noun, one referring to a concrete or abstract referent without discrete shape or number. In the (b) group, the same words appear but are used as count nouns, nouns referring to a concrete or abstract referent that has a discrete shape and can be counted: one song, two songs', one book,! wo books. Depending on what kind of linguistic theory we adopt, throughout its history English has either (a) added [+count] meanings to many [-count] nouns (and vice versa), or (b) has shifted a word from the f+counL] category to the [-count] category (and vice versa). These nouns, for example, were at one time purely count nouns: rock, stone, cake, pear!, brain, Hrer. In the course of time, they have also come to name the material of the objects they originally referred to, thereby becoming [-count] as well as f+count]. On the other hand, these next words were originally [—count] nouns, but are now both [ f count] and (—count] nouns: brick, rice, wine, fruit, meat,paint, pain, thought. All can now be used in the plural: bricks, rices, wines, and so on. While this is in one sense a semantic change, it is also a change with syntactic consequences, for we add plural -s and use articles differently with c-ount and non-count nouns. PROBLEM 9.10: Here are some noun phrases to put into the frame / see How do the plural s and the articles a and the occur with these nouns? Is there a general rule you can formulate? (Assume that those phrases and words marked * are ungrammatica] when they occur in the indicated syntactic

frame.) *girl girls a girl dust *dusts *a dust *car cars a car chili *chilis •a chili

*a *a *a *a

girls dusts cars chilis

the the the the

girl the girls dust * the dusts car the cars chili *the chilis

PROBLEM 9.11: The object noun phrases in I see the girls and I see girts are, respectively, definite and indefinite. It is the lack of an overt article in / see

BETWEEN SEMANTICS AND GRAMMAR

227

girls that indicates the indefinite girls. We might account for this pattern with a rule such as the following: ART—> [ +definite]. We would then "spell" [—definite] as either ajan or as nothing, [~definite] as the. What conditions determine how [ — definite] is spelled? Now that we have explored some of the grey areas between dear-cut semantic problems and clear-cut grammatical problems, we can see that describing hisLorical change in this area must be done on a very abstract level. Tn the past, most historical linguists were satisfied to coltect the data and categorize the ways in which the data could be organized. But this does not get to some of the deeper problems of linguistic change, problems that depend in good pan, again, on the particular way we choose to describe our own living language. For once we discover the crucial categories and processes in our own grammars, elements such as causa tives and so on, we are compelled at least to try to describe dead languages in the same terms, an attempt that very quickly reveals how difficult it is to reconstruct the linguistic abilities of speakers who have been dead for over a thousand years.

Chapter 10 GRAMMATICAL CHANGE: THE NOUN PHRASE

We have seen how transformations can rearrange the major sentence elements in a deep structure to produce a variety of surface structures. Most of these transformations delete a subject or objecL, or both delete and rearrange: / saw (some unspecified object) -* I saw. {Some unspecified agent) cooked the meat -> The meat cooked. PROBLEM 10.1: How would a transformational grammar deal with the following sentences: 1. 2. 3. 4. 5. 6. 7.

Him Hike. Down the street came a truck. Those questions no one can answer. A cross the way was a targe dog. Happy you may be but lucky you are not. 7YJ grandmother 'л house we go I Then came an import an t part.

How would a descriptive grammar deal with them (review page 26ff) ? 229

230

OKAMMAR AND SOUND

WOHD ORDER: THE MAJOR ELEMENTS The deep structure of these sentences, of course, would be in a basic, specified order, and transformations would then shift their elements into the desired order; fl) I like him • > Him Hike and so on. Transformations of this kind do not create new subjects and objects. They are stylistic transformations allowed by the limiLcd freedom of ModE from the more usual subject-verb-fcomplement) order, fin complement, we include direct objects: I saw Tom; indirect objects: /gapeTom money; and elements after copuias like be, seem, become, appear, fee!, look: I am in the house, He became a king, You appear sick.) The freest elements are adverbial modifiers: I left yesterday -> Yesterday lief!, /usually sleep -> Usually/sleep. Statistically, however, the order of the main elements in ModE deviates very little from an S-V-(C) order. If the order did vary, if it did allow us any Sequence, we would have to rely heavily on semantic and contextual elues to understand which noun phrases were subjects and which objects. If a sentence like John saw the house could occur in any order: The house saw John, Saw (he house John, The house John saw, John the house saw, Saw John the house, of course, we could be fairly certain who was seeing what. But if we substituted bear for house, we would be less certain. And if we substituted George for bear, then depending on its context, the meaning might be very unclear. 10.2: In any ten consecutive sentences from a newspaper story, reverse the subjects and objects to determine whether a reader would be seriously confused if he couFd not tell from position alone which noun phrase was the subject and which the object in those sentences. Docs this suggest anything about the importance of word order ? PROBLEM

10.3: Here are some OE and ME sentence types that represent onlystrong tendencies, not invariable rules. What do you conclude about O t word order? About ME? What do you conclude about the range of variations among the main sentence elements and whether we can predict those variations by any principles of grammatical context4.' (The translations given in (IH15) and (16)^(20) are word for word.) PROBLEM

Old English 1. 2. 3. 4.

.

Se ntan is god. (The ma n is good.) Ne rnihfehegehealdan heardne mece. (Not might he hold grim sword.) pasetidesecyningpamdisc. (Then sent the ki ng t he d ish.) pa he pone cynfng sohle, fie heoiode. (When he the king visited, be boasted.)

GRAMMATICAL CHANGE: THE NOUN PHRASE

5. 6. 7. S. 9. 10. 11. 12. 13. 14. 15.

251

He hi fedan sceoide. (He t hem feed s ho u Id.) Ne drincs! fm win 1 (N о L dri п kcs 11 h ou w i ne 7) Nu ispeosgifu eow rtbraden, (Now is this gift (from) you taken.) Heo hine Ixrde, (S he h i m ad vised.) № hahba'6we cyning. (Not have we кing.) Hewear8w3t?tffo.{\\£\itcbrat;(b)vaaT man.) Godbeheadus p.ft we fist treowne hrepvdon. (God commanded us that we that tree noL touch.) Hewiesyltfe.(Hewasold.) We eow can. (We you know.) Pa com ntxssepnvst (Then came priest.) Hlhmfdcm hirametegeboht. (They had bought their food.)

Middle English 16. Nubiseche khpe, (Now beseech I tfiee.) 17, /hymfolwed.((hirafollowed,) IS. i ne can tie I tie ttiai fellen foe wander. (I not know nor I not may tell the wonder,) 19. pa pe King Stephne to Englalende com, {m macod he his gadering &t Oxeneford. (When King SLcphen to England came, then made he his gathering at Oxford.) 20. A leafdi wes mid hire fait bbet alabuten. (Л lady was with her foes beset all about.) 21. We redeth (read) i j>e holt godspelie (gospel) of to-dai pal ure (our) lord Jhesu Cast yede (went) one time into arte (a) snipe (ship) and hiss deciples iHid(y/itbi)himvit0pe see. 22. Noil рои hest (have) y-hyerdpe seanes (sins) pet eomep ofglolomye (gluttony) and of lecherie. Andpervorepet zuyche (such) zennes arizep comtHunliche ine tpverne, pet J.T weile (well) of zenne, Ьеггоге icit wylie a lite (little) take (touch) ofpezennespes byep (be) y-do (done) inepe teasefне. 23. Whanlherdehem,lwes ryght glad. 24. And whawe (when) he cons, and was nyj (near) pe hous, lie herds a symphonic and oper noise ofmynysir/vye (minstrelsy). 25. Alpe longnge (language) offre Norphumbres, and'speciatych at ^ork, ys so scharp, slyttym (piercing) and frotynge (grinding) and uftschape (unshapen), pat we Souperon men may put longage unaepe (with difficulty) undurstonde. From these sentences, we can see that word order ш ОЕ was considerably rnore varied than in ME or ModE. Some linguists have even claimed that OE word order was almost free, allowing virtually any order among the major sentence elements a* the demands of rhetorical emphasis demanded. But the order was not "free" in the sense that it was not in most cases, outside of

SOUND

poetry, rule governed, that it could not be predicted. Most patter us were highly regular: 1.

The usual word order was that of ModE: S- V-(C), He geseah pone mann. (He saw the man.) |>et Estland is swySe my eel. (Estonia is very large.)

2.

In many kinds of dependent clauses, the typical (though not invariable by any means) order was CON JU NCTION-£-(Q-V: J>eah he him lcof wwre, he weep. (Though lie (to) him dear were, he wept.)

3.

When an adverbial element lite the negative ne or the adverb pa (then) occurred initially, a very frequent order of elements with a single verb wasADV-V-S-(C): Ne mihte he gehealdan heardne mece. (Not might he hold grim sword.) |>a sende se суп ing pone disc. (Then sent the king tlie dish,)

4.

When the object of a verb was a pronoun, the typical order was S-O-V: He hi fedan sceolde. (He them feed should.) pa hurgware hit geftiemdon. (The citizens routed them,)

A transformational grammar would deal with these patterns through a series of obligatory transformations, transformations which (in opposition to those that would pnHiuce. a sequence like Him i like) were not stylistic, but governed by grammatical context. In tact, the failure to observe these transformations may very well have been in OF. prose a stylistic feature that would have been noted by an ОE reader or listener. (150,174,221) 10.4: Rewrite this passage of ModE prose observing these transformations. (1) OBJECT PRONOUN SHIFT; (2) VERB-ADVERB SHIFf (for negatives and J?a) (For this one, you will have to rewrite ModE negatives into OH ne negatives and then into }>a.) (3) SUBORDINATE CLAUSE VERB SHIFT. When a linguist studies the history of a language, he does not study it in isolation from its social surroundings, fie does not ignore the society it depends on. If he overlooks this social context, then he risks turning language into a bloodless formalism. Then the study of language becomes cut off from its reason for being, When we forget the roots of language, we art no longer sensitive to the interaction between man and man through language. Then we fail to recogni7e (he single most important cause of linguistic change. We may siudy il !or years, all the while overlooking therealsources of its evolution—the interaction of social class which brings into contact speakers fi"om different social levels. Although these same transformations can be found in Larly ME as well as OE, they began to occur much less frequently after the thirteenth century.

GRAMMATICAL CHANGE; THE NOUN PHRASE

235

Between Lhe years 1000 and 1500, For example,, the Frequency of an accusative object before a verb decreased markedly:

Accusative object before verb: Accusative object after verb:

moo 48%

120(1 53 47

T300 40 60

1400 [4

tsoo 2

In fact, by 1350 or so, the word order of main elements in written English was virtually that of Mod E. (28,65)

INI LECTION: OE, ME, AND ModE In order to understand why these larger patterns may have become more regular when they did and as they did, we have to examine Lin aspect of grammar we have touched on before. 10.5: Problem 4.5 distinguishes derivational affixes from inflections. The endings we shall define as inflections for ModE are these: 1. 2. 3. 4. 5. & 7.

The plural -s (or other sign of plurality: boys, feet, men, and so on) Th e ge ni ti ve -s (boy^s, boys', children's). The thi rd perso n singu la г -J (//<: walks; It stops). The past tense -ed of verbs (or other signal corresponding to it: We walked; He ran; They flew: I hurt it). The past participle -ed of verbs (or other signal corresponding to it: He has grown; She has mug-., Unite walked). The progressive partieiple -ing (Me was singing. Distinguish this from the suffix -ing that makes a noun out of a verb: Singing is fun). The comparative -er or superlative -est of adjectives (bigger-biggest).

To determine how important inflection be for ModE, take 250 consecutive word from a newspaper or magazine and deiete every inflection from it- All verb will be in it infinitive form; nil noun and pronoun in it singular and nominative {he, she, /, you, ti) form; all adjective in it base form. To what degree do the loss of inflection obscure communication? Assume you can get use to reading sentence without inflection in it, what semantic information be lose when inflection be iose? Do any syntactic pattern become ambiguous? Which inflection be need much? Which be unnecessary? Now do the same thing for all affix. Which be some crucial to English, inflect or suffix? What take the plate of inflect V Write you answer to this quest without any inflect. Then write it agдin but omit the aiilx.

234

fiRAMMAR

ANJ> SOUND

THE NOUN PHRASE: INFLECTIONS AND STRUCTURE Although OF, had an elaborate paradigm of inflectional endings for both noun phrases and verb phrases, it was Lhe inflectional system for the nouns and their associated adjectives and articles that is most relevant in explaining why О Е word order may bave changed as it did. The remnants и Г this system can be seen in our pronoun paradigm with its three LO four forms for most of the pronouns: / you he she it

myjmine yourjyours kisjhis herjhers Us!—

me you him

we you they

ouriours yourlyours theirjttteirs

us

you tkei

her it

The function of these different forms is LO specify how the noun phrases they occur in relate to the main verb, to the preposition, or to another noun. Given I saw hlmt Him I saw, I him saw, Saw I him, Saw him J, we could decide who saw whom, regardless of word order, if we were certain Lhat we could rely on the form of the pronouns to reveal how they related to their verb: him = object, / = subject. But in OE. not only pronouns, but nouns, articles, and adjectives displayed these distinctions. We could, for example, write se man geseah pone heagest (the man saw the horse), se man pone heagest geseah, pone hetigest geseah se man, and so on, and always know which noun phrase, se man or pone hettgest, was subject and which objccL. Sn was a masculine singular article that always marked the nominative case, makingsetnun unambiguously the subject, pone always marked the masculine singular accusative, making fionehengesJ an unambiguous object. Thus word order in OE was not always crucial to identifying subjects and objects, even though it was rule-governed, because in add! Li tin to word order and semanLic sense, OE had this third system thai specified grammatical relationships: inflections. It was a system that would almost disappear by the end of ME.

Case, ModE and OE Of the three distinctions this system displayed in OE—case, number, and gender—case is (and conLinues Lo be in ModF.) the most complicated. Only among Lhe pronouns in ModE do we still cross-classify by three overt cases: nominative, accusative, genitive. But when we investigate in the context nf an enLire sentence how Modli pronouns work, we discover that, in fact, cage is a

GRAMMATICAL CHANOE) THE NOUN PHRASE

235

necessary implicit feature even for common nouns; even though a noun may not be overtly marked for ease, our grammar must always mark it for potential case. Whenever we repeat a noun phrase within a sentence (or between sentences), we ordinarily have to transform one of the noun phrases, normally (though not aiways) the second one, into a pronoun. If we begin with a semantically explicit deep structure that we can roughly paraphrase, as Tom left when Tom saw Tom's girl walk by Топь we recognize that in Fully grammatical English the surface structure must be more like Tom left when be saw Iris girl walk by him. If we assume that we begin with the specific noun phrases and then transform them into the pronouns, then even in ModE it is clear that, we have to be able to specify potential case relationships for every noun because it might become a pronoun. Jf the noun is mentioned in the immediately preceding discourse, for example, it is often turned into a pronoun, which must be marked for nominative, accusative, or genitive case: he* him this, H ow we specify these cases in a grammar of Mod E is in some ways quite different from how it may have been specified in OE, a fact which indicates a.major change in the grammar. At least from the written evidence, in OE, the nominative, accusative, genitive, anti one more case which we have lost encirely in ModE—the dative case—'were determined in these two ways: by how the noun phrases they marked related grammatically to their governing verbs or prepositions, and by how those noun phrases functioned adverbially in their sentence. Noun phrases that were subjects or predicate mnmiials, regardless of their actual position in their sentence, were in the nominative case: ic hit eom (I it am), we hie simfon (we they are). Noun phrases Lhat were objects of verbs and prepositions took different cases, depending on the particular verb or preposition. Some verbs required objects in the accusative case: He afshg pone mann (he slew the man): pone was the accusative masculine singular definite article. Some required objects in the dative case: Hextwindefi p*m manne (he escapes (from) the man): fmrtn was the dative masculine singular article. And some verbs required objects in the genitive case: He ehte(tfwsw<mnes(he pursues the man). p
236

UKAMMAR AND SOUND

relationships generally took the dative ease; there were no prepositions that exclusively required the genitive case in their noun objects, but some occasionally were followed hy geniLives. This kind of case agreement requires oniy that in the dictionary, entries for prepositions rind verbs include the notation that the noun phrases following t.hcm muSL be in the required case. A transformation can then attach the appropriate case- to the article and noun. For example, in this sequenee, the feature [dative) is transferred from the verb to the ART and N. The sum of the features is then ''spelled " appropriately by another set of graphemic rules which we shall discuss briefly later: ART

N

P[+definite)

mann[ +masculine]

[-singular]

ART j [-1-definite] [dative]

N

N*

[ +masculine] [dative]

[—singular]

N

I

he [dative] N

he (he

V

aetwindeS escapes

I

1 "I:: "I: I: -1

the

men)

I N " refers to hl number," singular {+singular], or plural [—singular].

10.6: Is there any equally simple way to specify which nouns should be in the nominative case?

PROBLEM

The second principle by whiuh ease was assigned Lo noun phrases depended on how those noun phrases functioned in their sentences. Noun phrases that functioned as adverbs indicating an extent of time or space, for example, often took the accusative: pa sxton hie fwme winter set Cantwarbyrig. Then spent they the winter at Canterbury. Adverbs of instrumentality or time often took the dative (-am was a dative plural ending). Hwilum incs he god. (At) times he was good.

He u'.i'MviuulLim werig.

He was (by means of) wounds weary.

GRAMMATICAL CHANGE: THE NOUN PHRASE

237

Adverbial phrases of rncasure took the genitive (-я was a genitive plural ending). Daga о nd nihta drincad we meodu. (Of) days a n < ! nights drink we mead. (Note the adverbial shift in these examples.) How a grammar would assign case for these patterns is more complicated than the means described previously. We must, of course, expand our base rules to account for these adverbial phrases: S^NP

VP

(ADVaoJ

(ADV B l a M )

(ADV m a M f t r ), . .

We might assign different case features directly to these adverbial units. It has also been suggested that in their deep structure, all adverbials are best treated as prepositional phrases. The details uf the argument need not concern us here, beyond pointing out that interrogative adverbials like when, why, hew, and where can be paraphrased as: at whet time, for what reason, in what manner, at what place. Since they arc synonymous, some transformationalists have argued that the more general, the more powerful explanation of such adverbials would be to begin with the preposition and then delete it where it does not occur in the surface structure. (55, but see 94 for the opposite view.) Such a proposal would greatly simplify how we assign case for OE adverbials, for we would then simply list the case feature fur the preposition in the dictionary as described previously, shift that case feature to the noun phrase object of the preposition, and then delete the preposition. For example, n in hwilum НЖУ he god ([At] times was he s ° d ) , hwitum with its -um ending is in the dative plural case. In the deep structure of this sentence, we might postulate a preposition that if realized in the surface structure would be spelled rl, a preposition that governs a dative object. After that dative feature is transferred to the noun phrase, the preposition would be deleted leaving only hwiium: Prep

N

№

set [dative]

hwtf[feminine]

[-singular]

->

«t

hwil[feminine] [dative]

[-singular]

->

#

hwil~ [ — singular] [feminine] [dative] hwil-um

238

GRAMMAR AND SOUND

PROBLEM J0.7: Let us assume that comparative constructions like the following result from an ellipsis of a fuller deep structure: He is taller than George is tall. -y He is taller than George (is). If we do delete elements, must we be certain that in OE, the nouns following f>c?rme (than) are in the correct case, the nominative case, before or after such a transformation'.' How does this compare with casual educated spoken ModL? With formal wriLlcn ModE? Here is an OE sentence to use as an example:

He was better than I. PROBLEM 10.8: From the few examples you have seen of the dative and from these next few word for word translations, how has ModE compensated for the loss of the dative case? How would you explain the change transformationally? descriptively1.' (The quotes run on separate half-lines.) 1.

Eadmwids3e!mg, ealdoriangne tir / geslogon %t sxece sweorda еcgum,,, Edmund (the) nobleman, lifelong glory / won at battle [with] swords' edges.

2 3, 4,

. . . so [to] him noble was/ from ancestors. . . . . . feMdznnetkIsecgaswtfi, ., - ., field screamed [witItJ warriors' sweat (i.e., blood) pstr kegseegtttenig { «aTnmforgrnnden .. . There lay warrior many / [hyj spears killed . . .

(ecgum: dative plural; him: dative masculine singular; swale: dative singular; gar urn: dative plural.) How would one decide whether, in principle, it were better for a language to have inflections oi1 periphrastic constructions to express the kinds of syntactic relationships suggested here?(177)

Gender, Modi: and OE Another, though, much less important, difference between OE and ModE is that OE formally distinguished grammatical gender in both nouns and third-person singular pronouns, while ModE signals natural or conventional gender only among third person singular pronouns: kethim,tiig;she,her, hers; it, U, its. We distinguish these ModE pronouns, of course, by whether we refer to naturally masculine humans: boy, man, priest, waiter; naturally

GRAMMATICAL CHANGE! THE KOUN PHRASE

239

feminine humans: wvman, girl, waitress, priestess; or to nonhurrians or naturally inanimate objects: rock, book, desk, idea, pig, dog, chicken. But we have also conventionalized certain referents as feminine: She's an old tub, but a seaworthy one. Hurricane Alice blew herself out. ., In OLi, on the other hand, grammatically masculine, feminine, and neuter nouns did not always correlate with natural gender. Among the masculine nouns were wtfmatm (woman), stan (stone), aS (oath), bat (boat), hfaf (loafJ, mete (food). Among the neuters were mxgden (girl), scip (ship), Itm (limb), ban (bone), bedd (bed), wedd (pledge), spere (spear). Among the feminine were glof (glove), ecg (edge), hwit (space of time), ides (woman), iniiu (might), r;
When I threw the stone, he (referring to stone) hit the window, Ttien [picked it up add threw it again. The maiden helped itself (referring to maiden reflex ively) to the food, but later she put it back. The glove was too small so I returned her (referring to ghee) to the store. They wotddn't take it hack, though, The woman loved the man who kissed him (referring to woman) but she didn't marry him.

Number, ModE and OE In Problem 9 JO we began to deal with the problem of number through the semantic-grammatical change that shifts a word like grass from a category of nouns that does not occur in the plural to a category Lhat does. The syntactic consequences of Lhis uhiinge are that such nouns may occur not only with the plural -5 marker, but after the indefinite article а/аи, as well: a grass, several grasses. The [-count] sense of grass does not take a plural and may occur without any article at all: Grass grows everywhere. Contrast *Car drives everywhere. In order to account for these distributions in ModE and to understand how OE noun phrases have changed in regard to number, we have to sketch— very briefly -how we might grammatically describe ModE number. We already have the single base rule for the N P : N P ->ARTN. Because every M P must be eiLher singular с г plural, we will add to this rule a symbol we have already , 1 * лгеапкап un^rammaiica] construction.

240

URAMMiR

AND SOU МП

used, number, N": NP > ART N № . We can then add another rule: № - y [ ±singular]. This will allow us to specify whether the noun will be singular (E-h singular]' or plural ([-singular]). We need this not only to account for plurality in nouns: boy-boys, and so on, buL also to account for the problems involved with the distribution of aicm and the. The definite article may occur with any kind of noun, singular or plural, count or non-count: the boy, the boys, the food- The indclinile article я/ян may occur only with count nouns in the singular: a boy\ not *a boys or *я <:haa$. Indefinite count nouns in the plural may occur without any article: Bays will be boys, as may indefinite non-count nouns: Foodmeans strength. In order to make sure these features of count vs, поп-count, singular vs. plural, and definite vs. indefinite are indicated in all of these phrases, we have to mark every noun phrase for [± count], [ i singular], and [± definite]. Now, since noun phrases that have no articles are [-definite], that is, general, unspecified nouns, it makes sense to say that in the deep structure of phrases like boys in Soys will be boys пай food in Food is goodfor you there is in fact a [-definite] ARTICLE, even though it is not pronounced in the surface structure. One simple way to accounL for this is to make the ARTICLE an obligatory choice in the base rule: NP -^ ART N N"and then add a rub for ART: A R T - * [ + definite]. We then have yet another kind of transformational rule, not one that shifts elements about or deletes them, but spells certain groups of semantic features into words. For example, consider these three trees: (I)

(2) NP ART

[+definite]

NP

N

N"

ART

boy [-singular] [+count]

[—definite]

N

№

horse

[ — singular]

[

(3)

ART

I [—definite]

grass [-bsingular] [-count]

First, we nave to transfer the features of [f count] and [ I singular] from NOUN and NUMBER to the ARTICLE, since the ARTICLE must «Sect

GRAMMATICAL CHANGE : THE NOUN PHRASH

241

whether the noun is court лг non-count and singular or non-singular, and NUMBER to the NOUN, since its spelling will reflect number too. Here is what happens with (1), ART 1

[ +definite]

ART

N

№

1

[—singular

i

boy [ + count]

i

1

N

1

i

i

[-?-definite] [ +count] [-singular]

boy [ + count] [—singular]

Then a spelling rule spells the ARTICLE in the appropriate way. Tn this case, the ARTICLE is spelled the and the NOUN, because it is [-singular], is spelled bays. 10.10: Spell trees (2) and (3). What is to be done with proper nouns like Gevrge in George is here ? The particular expression of definite and indefinite reference for noun phrases in OR differed in many respects from our own. Often, no overt article at alt was necessary where we would have to use eiLher a definite- the or an indefinite a I an: L

2,

3.

In devise ahbudissen mywstre нж? sum brodor syndrigtice mid godcundre gife genwred.. . In this abbess's monastery was a certain brother especially with divine gift famed . . , От! far his leousongum monigra monna mod oft to woruide forhogdnisse. . . onbxrndewn'tan. And for his poemsongs many (of) men's minds often to world's contempt, . . inspired were. SeCyneheardwms }>xs Sigebryhtes Ьгодог. The Cyneheard was this Sigebryht's brother.

In (1), the si ngular gift is the first ment ion of the ©ft in Bed e' s story of the poet Csedmon. We should translate it as, . . with a dlofaegift. Conversely in (2), we now require the world instead of just world. On the other hand, OE could also use the definite article where we ordinarily would not- -before names as in (3).

242

GRAMMAR AND SOUND

TABLE 10.1

PARADIGMS FOR OE ARTICLES, NOUNS, AND TRONOLNS Singular Article

1. Masculine; (stone)

Nom.: se Ace.: pone Dat.: piern Gen.: psss 2, Neuter: (ship) Nom.: pst

Ace,; fat Dat.: p&tn Gen,: pses 3. Feminine; (tale) Nom.;jeo Ace.: pa Dat.: pxre Gen.: pxre 4. -an Group (mixed Nom.; (same Ace.: articles Dat.: as Gen.: above)

Shared element

Noun ending

Pronoun

Inierrog.

-0

he hine him

hwa hwone hw&m

—

his

hwiES

-s

hit Mt him his

hwiet hwxt hwxm htvses

-t

{se staii) (pone stun) -e (jjxm stant?) -es (Jxts sianes) •Q

•0

(pset scip)

-0 (pset scip)

(px/n scipe) -es (jyxs stipes) -L'

•ne

-t -m

-s

heo (no — hi feminine -e hire forms) (ржге Ше) -(r)e hire -e (Jxzre tale) masculine, feminine, neuter nouns); (name, heart, ear) -я/-е/-е (se nanta, Seo heorte, px! eare) -anl-anj-e (pone патан, seo hear Ian, pxt eare) -an {psm name», t bxre heortatupxm earan) -an {pxs mirnan, рагге heortan, pss earan)

-0 (seo tab) -e (pa tale)

•('>

On the other hand, there was no OE indefinite article correspondirg to ModF. e/ви at all. Otir modern a/on developed in ME out of the number an (one) as it phonologically and semanticidiy weakened. The closest thing to an indefinite article is sum, which is less definite thin the definite article but more definite than the abscnecof any article, which communicated the more specific sense of indefiniteness. Overall, the paradigms for the OE articles, nouns, and pronouns, both interrogative and personal looked like Table 10.J, At the far left is the shared element that characterizes the particular gender, ease, and number. Groups (4) and (&) are examples of some of the more common irregular nouns in OE. (25,28,148, 174) 10.1]: Is there any opposition between cases that is maintained in every gender or every Dumber? That is, masculine and neuter dative articles and endings, pxm and -e, are identicaL erasing any gender distinction in that case. In the feminine, the dative and genitive articles and endings, part, and

GRAMMATICAL CHANGE: THL NOUN ^HKASK

243

TABLE ЮЛ (Continued) Plurals Article 5. Masculine Nom.:/и Ace.: pa Dat.: $'£in Gen.: £<м 6. Neuter Nom,: (same Лес.: as Dat.: above) Gen.: 7. Feminine Nom.: (same Ace.: as Dat,: above) Gen.: -an Group Nom.: (same Лес,: as Dm.: above) Gen.:

Noun ending -as

(pa stanas)

-as

[pa st&mtis)

-ига

(pmrn stanum} (pora siana)

-fl

(pascipu) -w/0 (pascipu) -urn (рхт sctpuffi) -a (parascipa) •Ф

Pronoun

Interrog.

hie

(no plural forms)

him hira (same as above)

Shared element — — -m -(ar)a — — •m

-(ar)a

-a (fxitala) -a (pa tala) -urn (J)£tn tahtm) -a ifraratalo)

-a -a -m -(ar)a

-an (pa namant earon, heortan) -an (pa naman, earan , /icorlan) -uin (pxm natnunj, eantm, keortum) -ena (para nammia, earenti, keoriena)

— — -m -a

-e are identical, neutralizing the case distinction in that gender. How crucial were gender and case then 1 10.12: ModC pronouns reneel gender only in third-person singular. Create a pronoun system wbicb wonld be maximally distinguished for gender and number in every person. What would be the eonsequences'.' Create a gender system that would be minimally differentiated. What would be the consequences'? Can these two systems be combined? H
244

SOUND

A second paradigm of articles was available alongside the one just sketched. IKHll.

muse. singular шип. ace.

dat. gen.

pts

pisne pissum pisses

fem.

pis pis pissum

pi'OS

pisses

pi.tae

pas pisse

plural norn. acc. dat. gen.

pas

pas

pissum pissa

The dilTcnciicc between them is at times not always clear from the context in which they occur. Perhaps they can be besf understood if we imagine our own article system reduced from the three-way opposition of ike—(this! these)—{that* those) to a two-way opposition, with ihatjthose absorbing the function of the. Adjectives as well as articles had to be iniiicLcd to agree with gender, case, and number of Lhc noun. The agreement system, however, was further complicated by the fact thai OE distinguished weak and strong adjectives, much as Modern German does. The principle was this: The distinguishing form for case, gender, and number usually had to mark the first clement in the noun phrase and the noun. If the first element was an article (se man), then the article, «?, had to indicate by its form its gender, case, and number. If the article was absent and the first element was an adjective, god man (good man), then the adjective has to carry an inflection (or a distinctive lack thereof) almost identical wilh the ending of an article. In such casts, the adjective was strung because it signalled gender, case, and number: tmmum /ahum (firm people): neuter, dative, plural When lhc adjective followed the article:/wm tinman foleum, then the article made the distinctions and the adjective was weak, taking a different and much less differentiated set of endings.

FROM OE TO ME TO MudE

Inflections Long before the end of the ME period, by the thirteenth century, these inflections had considerably Leveled. Among the nouns, grammatical gender

GRAMMATICAL CRANGE: THE NOUN FHRASE

245

distinctions had entirely disappeared, along with all noun case markings, except the genitive. Below is a schema of the M E article, noun endings, personal pronoun and interrogative pronoun «hat corresponds to Ihc OE nominal system outlined in Table 10.1. With the exception of the feminine pronouns and the third person plural pronouns, we will use the Northern dialect forms (sec Figure 10.1), which eventually spread throughout England. Even by very early M L , the weak-strong distinction among adjectives applied only to single syllable adjectives that ended in a consonant, like god (good), not to adjectives likc_/>e (generous), open (open), or grew (green). Before the end of the twelfth century, however, as the strong-weak. distinction began to disappear, scribes were adding final -e's to adjectives almost indiscriminately. Some traces of case-gender distinctions can be found in Southern and Midland texts as late as the twelfth century, but disappear soon after that, (98, 150, 154) The following paradigm does not suggest the variety of forms found even within a single dialect. Over time, the second personal plural pronoun, for example, varied over ew, он, jew, ^ou, and Between OE and ME, the first and second person pronouns changed less drastically than the rest of the system. The earliest ОЕ texts had a distinctive

\NORTHERN

V

Л BAST MitlLAND

i I

—'

SOUTH WESTERW

/

I J

EASrrtiKN

Figure J0,1. Middle EngMsh Dialctt Areas

246

GRAMMAR AND SOUND

TABLR 10.2

PARADIGMS FOR MR ARTICLES, NOUNS, AMU PRONOUNS

Singular Article Weak Adj. Strong Adj. Nuitn 0 (man) Masc. Nom. the -eigode) 0 {god) 0 (man) Ace. Dat. U) {man(e)) •es Ate. (manes) Neut. Nom. Ace. Dat. Gen. Fein. Nom. Aci. Dat. Gen.

Pronoun he hine him

inter.

his

whm

№

what what whom whos who whom whom whos

(A)fl him his heolschejscho hir hir

bif(e)

who

whom whom

(Note: heo is the Southern and Western form, sche the Midland form, and srhct the Northern form. The third person plural pronouns below arc in the same order: roughly Southern, Midland, Northern. There was no distinctive interrogative plural.)

Nom.

the

-e(gvde)

-e(g(ttle)

Ace. Gen,

-es [stones) her{e)jherelpetr{e)

accusative singular form, but by late OE, the dative and accusative were identical. The forms listed below separated by slashes; meefme, are earlier and later forms in the OF and ME periods. The mia-mi, thin-thy distinctions in ME were contextual!}1 determined forms, a question we shall return to.

Nom. Асе. Out. Gen. Nom. Ace. Dat. Gen.

Singular ME Otichll meojnte me me me min min-mi pu peejpe

ihou iftee thee tkin~thy

Plural ME OE we we us, on usicjus us, аи us ur(e) ure ge ye eowicjeow eujyou eow eujyou eower

GRAMMATICAL CHANGE: THE NOUN PHRASE

247

In addition to these forms, OE and very early ME had dual forms, wit h and gir, which meant *we two" and "yc two." They were inflected much like the other forms: wit-uncirSunc-uttc-^uncer, git-iitcitlinc-uic-incer. PROBLEM 10.13: What are the major distinctions lost in the pronouns between ME and early Modi:? Here arc the early ModF (post 1500 or so) pronouns. (It should be noted that in many British dialect areas, ihou,thee,thy,thine, and ye Bit still used,) Plural

he

me my you your him his

it

it

she

her her

I you

itsjit

mine yours !your{e)n hisjhis(e}n

we

you

us you

they

them

our

your their

ourjour(e)n yours jyour{e)n theirsjtheir(e)n

—

hen-Shir (e)n

Our{e)n, yovr(e)n, his{e)n, ker(e)n, and their(e)n are from the Midland and Southern dialects. The forms ours, yours, his, hers, and theirs arc of North and North Midland origin. PROBLEM 10.14: The source of the modern myjmitie distinction is illustrated in the following ME and Early ModE examples. Ex plain what happened. But first review the OE genitive forms, my rfere here, mynherte, thisynne and thyn ojjfcnce, to slay mine enemies and ayde my friends, I cut mine arm, and with my Mood Assume my soul Shakespeare has these: Give every та» thine eare but few thy voyce, myowne life, mine uncle, my hand, thy foot, thy ambition. By the early eighLeenth century, my, and но on, is normal everywhere before nouns; rrme, and so on, as a pronominal possessive, PROBLEM 10.15: In rapid pronunciation, we say / know'm to mean either I know them or / know him. Yet when we rapidly pronounce I know their house or I know they left, we do not drop the initial th sound. Might there be an historical explanation for this? PROBLEM 10.16: Yours, hers, ours, and so on did not occur in OE. How would you explain their origin? They began lo appear in Early ME, eventually replacing the uninfiected anaphoric pronoun by the middle of ihe sixteenth century. One of the last examples of Lhe uninfiected anaphoric pronouns is a sentence Like . , - elernail dampnacyon isyoure. PROBLEM 10.17: His{e)n, our<e)n, your(e)n, her{e)n, their(e)n developed in the South and Midland of England. What might explain their introduction? What social distinctions do they make today in Lhis country? Where arc Lhey found? Why?

24$

GRAMMAR AND SOUND

10.18: Let us assume from the extant documents that OE based its case system strictly on the kinds of grammatical relationships described before and after Problem 10.6. That is, no matter where in a sentence a pronoun happened to occur, it was ordinarily in the case required by the governing verh or preposition. Here are some examples of ModE to consider. Comment, ft'sme. He's better than me. Him and me left. Him I dot?t like, Who did you want to see? With whan} did you wish to speak? You saw who? It is I. Whom didyou see? Between you and I, he's afetoL PROBLEM

Between OE and ME, the system of articles before nouns also changed greatly. First, an indefinite article ajan developed out of the number an (one), though its use was not as predictable as it has become in MndE. Second, the invariable/>f developed very rapidly in the twelfth century as the replacement for the entire article system of OE. Although some remnants of gender-case distinctions can be found among articles in the South into the thirteenth century, pa had become the standard article in Lhe North by 1200, (209) PROBU^

10.19; Where did fa, ancestor of the, come from?

The outlines of our demonstrative system had also developed. Ft is fairly evident that this descended from /i/r, neuter nominative; and that from pser, also neuter nominative, but from the other paradigm. From pes had come these, white those probably descended from the plural pa, which by analogy developed the plural pas. it is not entirely clear why the article system should have developed a near-far distinction. It may be that here-there, now-then, thusso provided some impetus. Although the ttioujtheejihy pronouns are still used among some religious groups in this country and in many parts of England, they had generally disappeared from educated usage at least by the beginning of the eighteenth century. But between their simple use as singular second person pronouns at the beginning of the MF period and their loss 600 years later, the particular use afiheeithott was part of a socio-iinguistic event that is still felt in most of the languages of Europe. During the fourth century A.D., the second person plural in Latin, rax, developed a second meaning. It began to be used when addressing social superiors as a mark of respect. It may have had its origin when the Roman Empire was governed by two emperors, <jne in Rome, the other in Constantinople. Those addressing Lhe Emperor as an institution were required to use the plural. Perhaps those in court personally addressing one of them also felt obliged to use the plural form. In any event, the practice spread across Europe and in the thirteenth century, to England, where in imitation of the upper-class French iujvous, the English began Seeing and youing one another. This usage soon included another dimension of social distinction. As the you became the mark of respect from inferior to superior or between socially

GRAMMATICAL CHANGE: THE NOUN PHRASE

249

superior equals, the thvu form came to mark intimacy between Family members and lovers, then developed as a mark of solidarity between speakers who shared a set of values. It also developed one more dimension of usage— contempt. In Shakespeare's Twelfth Night, when Sir Toby Belch is urging on Sir Andrew Aguecheck to write a challenging letter to the disguised Viola, he says: Go, write it in a martial hand; be curst and bricfc; it is no matter how wittie, so it bet eloquent and full of invention: :aunt him with the license of In къ: if thou thou'st him some thrice, it shall not be aniisse-. IH.ii. PROBLEM 10.20: Discuss the use of youfthou in one of the following: (I) A fairly long scene from Shakespeare (some suggestions: Hamlet, Ffl.iv; As You Like It, IV-i; Antony and Cleopatra, V.ii; The Merchant of Venice, IV.i; The Two Gentlemen of Verona, V.iii, (2) The links (including Prologues) between several of Chaucer's Canterbury Tales. (3) Sir Gawain and the Green Knight (particularly the encounters between Gawain. and the l*ady). Do not use modernized texts for the last two. (20, 21, 24, 51, 105, 155, 232) PROBLEM 10.2i: We arc no longer able to manipulate the social tone of our conversations through pronouns. What devices are available today to express social distance, solidarity, contempt, or close emotional relationship4,' Do not include tone of voice, posture, expression, and so on. Consider modes of address (what do you call your instructor 7 what does he call you ?), vocabulary, pronunciation, syntax. Write about thirLy lines of dialogue which begin in a very intimate tone of solidarity, switch to contempt about line JO, then to a superior-inferior relationship at about line 20.

Relative Clauses The rule for the noun phrase: N P ^ ART N №, is still too simple to account for longer, more complex noun phrases. Nor have we adequately dealt with the genitive. For example, these relative clauses would also have to be included in any description of the noun phrase: ike ww» who was dead, the girl whom I saw, the man whose friend left, the book which 1 read, the dog thai was by the door, the tree which is blowing in the wind, and so on. In those cases where a full relative clause is present, the relative clause has Lhc same basic structure as a sentence. The relative pronoun stands for an element in its clause semantically identical to the noun it follows. In the noun phrase, the boy who left, who left has a subject who and a verb left Who "stands for" or, more accurately, replaces the boy in the clause. If this is so, then the deep structure of a noun phrase with a relative clause would look

250

GRAMMAR AND SOUND

something like this: [the boy [the boy te/fJsW A transformation moves the repeated article and noun in the clause, the boy, to the beginning of its clause. Id the case of the boy [(lie boy left.], they are already there. But they must be moved in the boy [I ssw the boy), -¥ the boy [the boy i saw]. The repeated noun phrase is then transformed into a relative pronoun: the boy [the boy /лети-1] -»• the boy [whom fsaw]. In the case of a genitive, the genitive and its following noun are moved: the boy [I bought tbc boy's book] -> the boy [the hoy's book T bought] —5- the boy [whose book 1 bought] So we must revise the NP rule once more and add some transformations. Since a relative clause is only a transformed sentence embedded in a noun phrase, we can simply add an S to the NP: NF*->ART N Mu (S). The transformations (I) shift the repeated NP forward to the head of its clause and (2) change it to a relative pronoun. JO.22: Here are some additional data regarding ModE relative clauses. Comment. I• I. 3. 4. 5. 6. 7. 8. 9. H>. I1. 12. 13,

The man who left was в friend. The man whom i saw was a friend, The man that Isa w was и friend. The man that left was a friend. *The man wh ich left was a friend. The watt to wham I spoke was a friend, The man wiio(/n)f spoke to was a friend, The fttttn that [spake to was a friend. *The man to that! spoke was a friend. The man I spoke to was a friend. The ma» Isaw was a friend. *J talked to the man walked down the street. (I.e., the man who walked. . .) *The book who I read was dull

10.23: J-fL-ге are several examples of relative clause patterns from OE, MF, and Early ModE written in ModE, except for the relative pronoun. Sketch a very rough history of the changes in relative clauses and suggest the kinds of transformational changes that have occurred in the last 1000 years. If a form of a relative clause is asterisked, assume that it would have been ungrammatical in the period; that is, an example of it has never been found. For example, *Ic knows the man who iefte would not have been possible in 1300. (In the OE examples, jbe is an indeclinable particle.)

CHANGE: THE NOUN PHRASE

251

Ofc L 2. 3. 4. 5. 6. 7.

The man se was there left. The man pone I know stayed. The ship pxt was there sank. The spear pxt you bad broke. The щ\оие seo was (here was lost. The talu pa you heard is old. The maiden to psem f spoke left. The king topsem I spoke left. The w i/e tepicrelspoke left. The man pel saw left. The ship pe was there sank .The glovefiethe man looked for was lout. 8. The boat sepe sank was old. 1have the horse pmtefieyou want. The landpset pe is fertile is eastward, i saw the child pmm pe you entertained. I saw the woman fie left. The glove soe pe ripped is mine. 9. *The man pxm I spoke to left. *The ship pxm I looked at sank. *The woman pxre I spoke to left. 10. *The man to fie 1 spoke left. *The ship at pe I looked sank. *The woman to pe /spoke left. 11. The man I saw left. I saw the man walked down the street. I hate the sword к Hied the warrior. ME(prt>1450) 12. 13. 14 15, 16, 17, IS. 19. 20. 21. 22. 25. 24, 25, 26,

The man pel saw was here (frequent), The man to which I spoke was here (less frequent). The wan whom I saw 1ф (less frequent). The maul saw left. 1 talked io the man walked down the street. The man fixt I saw left. *The man to pxt I spoke left. The тип whose book I read left. The man who was here left (very in fneque tit), The man whom that you saw left. The man whose that book I read left. The man the which was here left. The man the which man was here left. The man which that was here left. I spoke to a man, which man was my friend.

ME-Eariy ModE (1450 -1550) (13), (14), {20) become more frequent but disappear by 1500; (21), (22), (23), (24), (25), (26) become less frequent. Early ModE (13) beeomes less common; (21), (22), (23), (24), (25) disappear; (2rt), The man which 1 saw left begins to disappear. (36,221)

252

GRAMMAR AMD SOUND

The Genitive

Constructions like Tom's hat, myfriend's defeat, the teacher's performance, Bill's weakness are closely related to the relative clause construction. But we have to distinguish four kinds of genitives, only one of which we can correctly call possessive: George's haf. We might generate such a phrase in a series of transformations roughly like the following:

-> -* -* ->

the hat [George ^possesses) the hat] the hat [the hat George (possesses)] that hat [which George possesses)] the hat [of George] George's hat

That is, in the deep structure of the noun phrase, George's hat, is an embedded sentence which defines the semantic relationship between George and hat, a relationship that eventually will be realized as the genitive; or possessive: George's hah (Whether the actual verb in the deep structure sentence is a specific word like have or possess or own, or whether the verb should be represented rather as an abstract bundle of semantic features that abstractly represents a deep structure meaning (if possession has been debated at length and generally resolved in favor of the more abstract interpretation.) 10,24: Comment on constructions like these: George's performance, ihe man's escape, my disappearance, your release, his defeat, their encirclement by the enemy, our capture, her sickness, Tom's weakness, his intelligence, their strength. (Their equivalents were found in OE.) PROBLLM

10.25: Here are some genitives from before 1600 or so. Comment.

1. for the wyveslove of Bathe (i.e., for the Wife of Bath's love), 2. The Dukes wyfe ofTintagail (i.e., the D uke of Ti ntagail'a wi fe), 3. The Archbishop's Grace of York (i.e., the Archbishop of York's grace). This pattern ^was the rule before the 16th century. After, group genitives like the Duke of York's land became common, replacing the earlier pattern. (46) pRUBi.EM 10.26: In Early Mod£, patterns such as the man his heart was broken begin to appear. By the seventeenth century, the apostrophe possessive: the man's heart was broken is felt to represent a contraction. Comment. One very important difference between OE and ModE genitive constructions is that OE did not, ordinarily, allow the of- paraphrase: George's performance = the performance of George, his hat = a hat of his, the enemy's

GRAMMATICAL CHANCE; THE NOUN PHRASE

253

weakness — Jiie weakness of the enemy. Moreover, the genitive could either precede or follow the modified word: £ ш ejmtoges moan (the king's man) as opposed to sumne dselpass htafes (some part (of) the bread), Then, between the years 900 and 1250 or so, a major change occurred in the form of the genitive, a change related to that which fixed the S-V-(C) word order and increased the structural load carried by prepositional phrases. In 900, the genitive occurred before the noun about half the time: hisfeonda stege (his enemies' defeat), and after about half the time: <em~ge pinga ([by] any [of] means). By 1000, the inflected genitive occurred after its noun only about 31 percent of the time: by 11Q0, about 23 percent; by 1200 about 13 percent, and by 1250, less than 1 percent. By 1300, the inflected genitive was permanently fixed before the noun it modified. But as the genitive became fixed before the noun, the periphrastic genitive, the genitive expressed by of, began to increase in frequency. In Ш), the (^/-construction occurred in less than one half of one percent of the genitive constructions. By 1200, it still occurred in only about 6 percent of the constructions. But then by about 1250, it jumped to about 32 percent of all genitive constructions, and then to almost 85 percent in 1300. In other words, as the position of the inflected genitive became fixed before the noun, its relative frequency decreased in favor of the prepositional phrase after the noun. (65) The change can be best seen on a graph (Figure 10.2). PROBLEM 10.27: If Baugh's figures on the proportion of French words borrowed at various periods in ME are reliable (Problem 4rJ2), would they tend te confirm or contradict the claim that the of + NP phrase resulted from the influence of the French de + NP? The periphrastic genitive may have spread so rapidly because prepositional phrases were in general becoming more important. It may also be that J00

100 DO M:

^—-

Pre-noun inflected

70

~

Kfl

~

60

JO 40

\

/

\

1

—-^^

v> 20 10 0 SOD

.•-

•~~

—

, • "

•

=

"

—

—

-

-

.

isi

LOSO

ПНЮ

Л

\ \

= ~

I MO

Ml

/

[200

10.2, Construction of the Genitive

L25O

[0 0 L300

254

^ЙАЧУЛК AN» SOUND

the French genitive construction with d? became a strong model: Femme de chambre. Whatever its ultimate cause, ihe change further illustrates the two basic structural principles of our language after the Norman Conquest: |, 2.

Word order w;ts becoming increasingly fixed. I nflcctions were decreasing in Favor of periphrasis.

There is some evidence that even the inflected genitive is disappearing in some nonstandard. dialects today. Both in Great Britain and in the United States, one can hear genitive phrases such as my wife father, she boy hdtt and so on. Whether this inflection began disappear ing centuries ago and is merely reflected in the current speech of nonslandard speakers, or whether (his is a relatively recent innovation that heralds the loss of the last overt case distinction among common nouns remains to be seen.

CAUSES OF GRAMMATICAL CHANGI-".

Now that we have described some of the overl patterns of OF, ME, and ModF, noun phrases and have a very general idea of how the grammar of these patterns has changed, we can turn our attention to why these patterns may have changed in these ways. Like so many other changes in the language, (he replacement of a highly inflected, variable word order in OL by a slightly inflected, relatively fixed word order system in Mi: undoubtedly had. both internal and external causes. PRoni.F.vr I0.2K: This general change, the loss of inflections and the fixing of S-V-(C) word order, occurred between roughly 1000 and 1400, though the tendency toward this change can be delected as early as 900 in some dialects. The crucial period seems to be between about 11П0 and ООП. Looking back at the external history of language during these 4Ш years, and more particularly at the 200 years between П0О and П00, speculate what causes might have contributed to Mich a change. Whitt information would you particularly like to have7 What do the following data suggest? In ninth century OF. prose, subject and object are inflectionally distinguished less than half the time. When they are not, 94 percent of the distances are in Subject-Object order. When they arc inficctionally distinguished, S-O order occurs 93 percent of the time. (Robert L Saitz. Functional Word Order in Old Lnylish SubjectOhjccr Patterns, LJnpub. diss., Madison, Wise., 1955. Quoted in Cassidy and Ringler)

GRAMMATICAL CHANGE: THE NOUN PHRASE

255

In conjunction with these external forces, perhaps in a necessary conjunction for the change (o occur, was a major Lnlernttl force. We have said little about how OE or ME was pronounced. But here we must point out one phonological characteristic of OL and ME that may have strongly influenced the loss of inflections, a cause thai probably worked in conjunction with an increasingly predictable S-V-(C) order, even in OE, Generally speaking, words of Germanic origin and Romance words that were completely assimilated into ME stressed the root or initial syllable, unless the first syllable was a prefix like ge- or Ы-, Thus we have words like OPe/i, After, ANtmal, CAStte, Final, ACcident, ALchemy, STAXdard, Un~ assimilated Romance words, mostly borrowed since Early ME, usually stress the last syllable: tnuCHWE.poLICE, SaTRINE.proTECT, comMEND, and so on. In rapid heavily-accented speech, unstressed final syllables tend to lose their distinctive vowel quality as they are reduced to a kind of indistinguishable mid-central vowel called schwa, a sound symbolized /в/. It is the sound of the first syllable of allow or aboie or the vowel of the second syllable of zebra or bolted. \\\ OE, however, the vowel in the end ings had to be clearly distinguished in speech. Otherwise, speakers could confuse endings like -es and -as; -tin, -ent and -on; -ы, -e, and -a. As early as the tenth century, speakers began to reshape the lightly stressed dative plural ending -um to -unt perhaps by analogy with the other -n endings. Whatever the cause, by Late OE -un had begun to occur as the new ending for many -um forms. PROBLEM 10.29: How would this affect the oppositions among the cases and genders? Would any distinctions be lost as long as the vowel in -un were still clearly pronounced ? (See Table 10. I,) PROBLEM 10,30: By the eleventh century OE scribes were often writing words earlier spelled like those among the first of each of the following pairs in a new way, like those among the second of each pair. What does this suggest? fiscas-fisces, bitori-biten, trumatt-trumen, gwttan-gumett, eagutn-eagem, dxda-

HRORI.FM 10.31: If-um endings had become -un, and if endings that had once had clearly pronounced vowels spelled , <e>, , and , became endings spelled <e> and probably pronounced /э/, how would the oppositions 4 among the cases, genders, and numbers be affected -' Refer to the list of endings in Table 10.1 and systematically change all -ttnt to -un, then all vowels in the endings to <e>, PROBLEM 1U.32: By Lhe end of the eleventh century, the final -n in inflectional endings began to disappear loo, leaving only a general ,'э/ sound, spelled <e),

256

GRAMMAR AND SOUND

in place of /en/, spelled <-en>. How would the loss of/n/ affect the distinctions among case, gender, and number? Refer again to the endings listed in Table 10.1. Change all -ett endings to just -e. PROBLEM 10ЗЪ: Strictly on the basis of these changes, what should the inflectional system for nouns have looted like by ME? Here is the actual noun paradigm that resulted:

Singular Plural

\om,/Acc. Stan stones

Dative (Infrt •qiR'nf)

Geniti

slon(e) stones

stones stones

Where are the unexpected inconsistencies? What has to be explained in a way other than by predictable phonological changes? PROBT.FM 10.34: Eventually, in Late ME, even the final unstressed /э/ largely disappeared {though scribes added final <-f) indiscriminately, evidence that the system was breaking down). How did the loss of /э/ affect the inflectional system? What would you say was the crucial point in this progressive loss of inflections? PROBLEM 10.35: From the meter of poetry, we can tell that words like jumped, kittges, stopped, dogges, and egges were often, two-syllable words. Examine some poetry between 13OO-J80O to determine when they consistently became single syllable words. (146, 147,144,150}

SOME MODERN SURVIVALS OF OE INFLECTIONS

Despite the fact that most of the gender, case, and number system in OE was leveled in ME, we can find some survivals of the system in the irregularities and anomalies of ModE inflections. PROBLEM 10.36: There is one group of nouns in English whose plurals frequently have no -л endings: I saw twenty sheep, eight swine, four deer? two moose, six buffalo, five caribou, seven fish, twelve haddock, six perch, four salmon, two trout, three bass, nine tuna, four mackerel, and so on A n ME, there were a few others: two horse, four neei (cattle), six folk. Swine, sheep, deer, folk, neet, and horse all descend from a class of neuter nouns whose plural ending was -u. Fish was a masculine with a regular plural, fiscas. Buffalo is a Latin word dating from 1588; caribou is French Canadian, 1774; moose is a

GRAMMATICAL CHANCE! THE KOUN PHRASE

257

Narragansett Indian word adopted in the United States after colonial times. The fish-words come from a variety of sources, many borrowed into ME from Old French. Comment. 10.37: In other phrases of measure, ModE nouns also lack a plural, though in this case, the nouns come from any OE gender: a tat foot pale, a five dollar bill, a four year old child, a five yard gain* a twenty minute wait, and so on. In parts of this country and in England one can also hear phrases like It's two foot long* He walkedfire mile, andsoon. In OE phrases that count or enumerate a noun, the noun was typically plural genitive: seofan rniia (seven (of) miles), tweofeta (two (of) feet). What happened? PROBLEM

10.38: In ME, the plural -en ending was common in the Southern dialects. But the -s plural ending common in the North and Midlands Anally pushed -en out of the language, leaving only three relics in ModE: childchildren (actually double plural, the original plural being cildru), ox-oxen, and the archaic brother-brethren. "Why might we have expected the Northern and Midland forms with ~s to replace the Southern -n forms in Late ME, rather than vice versa? Another small category of irregular plurals in English are those which indicate plurality by a vowel change: foot-feet, tooth-teeth, goose-geese, mouse-mice, louse-lice, man-men. The group was iarger in OE, Their origin goes back to the Germanic period, before any written records. The Germanic • form of geese was something like *g{n>siz. The Prc-OE form was probably close to *gosi. In "very early OE the plural -r ending caused the back vowel /o/ as in goat to mutate or umlaut:, to become first a rounded front vowel like German o, which then unrounded to become a front vowel {ej as in gate. When the -/ ending was lost, only ges (pronounced like gase) remained to contrast with gos (pronounced to rhyme with dose). 10.39: These nouns were in OE also umlaut plurals. Had analogy not put them into the regular noun class, how might we pronounce their p!ura!s today: oak, goat:,foe, cow, shroud, book? PROBLEM

PROBLLM 10.40: How do we explain these irregular plurals: banditti, stadia, foci, cherubim, formulae, gladioli, phenomena, data, opera, agenda, stamina, stimuli, curricula'!

Adjectives and Adverbs We have already touched on the case and gender endings of adjectives. Anothcr kind of ending is the comparative and superlative. OE comparisons were much like some of those of today: keardm (harder) for masculine nomina-

258

GRAMMAR AND SOUND

tive and accusative, heardre for feminine and neuter nominative and accusative, heardoit (hardest) for all genders. The common element -r and -st is retained today. But there were also significant differences. PROBLEM 10.41: The rule generating comparatives and superlatives in OE would generate forms like modcearigra (more troubled in heart), feohgifrost (most greedy for wealth), andgitfullkost (most intelligibly), mmhygdlgra (more imprudent), bMra (more joyful), blindosl (blindest), tndryhtenost (most noble), hatheorteat (most hot tempered). In educated ME, we find comparisons like more wylde (wilder), moreswete (sweeter), mosiegrecyus (most gracious), moostprofitable (most profitable), otder, kngar (longer), most mylde (mildest), mare pore (poorer), mylder, moste old, the most fairest damyseues, the more gretter king. In Late ME and Early ModE t we find more ruefullyr, pooriste, horryblestt merueyllest, thinner, famousest, gladdest, diligent'este, mehncholiest, ancknter, more nere, most tkyk, mootf abiecte (abject), most rifch, more plesani, more auntieut (ancient), moos! adieuturous^ more fressher, more diligenter, most worshipfuliest, most sharpest wit ted, more worser, more neerer, most cruellest, most fittest. From about the eighteenth century, we find comparison much as it is today. Comment. Adverbs in OE and early ME were formed by adding -e to an adjective: beorht {bx\^t\X)-beohrte (brightly). Our -!y ending comes from an adjective ending -Нее which weakened to -ly and was then reinterpreted as the mark of an adverb. The adjectives friendly, homely, earthty, heavenly, and so on still use that original ending, PROBLEM 10.42: Sentences like these are considered "ungraminatical" by many grammar teachers. He walked slow. She sang real sweet. He talks too loud. He got up pretty quick. Sentences very similar to these are not considered ungrarnmalical: Иг walks fast. I work hard. They ran far. Comment, using a historical mode of explanation. One expression in ModE He works nights, I work days, We live there summers, sounds like a modern plural noun serving an adverbial functionSome historians of the language would argue that such expressions are "really" singular genitives. OE had a construction that added the genitive inflection to the noun to give a construction with the sense of He works (of) night. From the point of view of a linguistically naive ModE speaker, though, a sentence like lam there weekends can only be interpreted as having a plural noun in an adverbial function. At some point in the past, the genitive was simply reinterpreted as a plural. From the same source comes the final -J of once from ones (ones), twice, thrice, homewards, southwards, sideways, always, sometimes, backwards.

Chapter 11 GRAMMATICAL CHANGE: THE VERB PHRASE

INFLECTIONS: MndE AND OE

Like the noun phrase in ModE, the verb phrase lias also lost most of an earlier, more extensive set of inflections. And like the noun phrase, the verb phrase has also developed an elaborate system of periphrastic or phrasal constructions that communicate fine shades of tense, aspect, and voice: He will have been helped, He has been being careful, and но on. We will examine the history of these patterns as we did the history of the noun phrase: First we will organize the data to see how the forms of OE, ME, and ModF. verbs differ. Then to understand how the grammars that generated these structures have changed, we will construct some very simple rules for them. Finally, we wij| speculate about what caused these changes. The ModE verbal inflections are limited: (I) to the third person singular present tense -s: He goes', (2) to a. general past tense inflection, often called the preterit, that we spell -ed for the regular form, in a variety of other ways for irregular verbs: sleep-slept, tkink-thought, felt-fell, go-went. Two other inflecLions indicate aspect, or the state of the action: ongoing, over, about to begin, and so on. (3) The progressive -wig form marks the verb after the progressive be and signals the ongoing but usually temporary nature of an 259

260

OHA4MAR *MD SOUMTi

action: Contrast Й is mining here with // rains here. And (4) the perfect participle which, like the past, marks verbs in a variety of v/sy&: fatten, seen, gone, stopped. 11 mar к s t he verb after the au x i Li а гу /щи? that sign aJs t be bou nded period of an action r He saw the picture vs. He has seen the picture, and the verb after the passive be: He was seen.

The Past and Perfect The ModE past tense and perfect forms arc difficult to describe briefly because of reasons that go back into (n do-European prehistory. A pattern that distinguishes Indo-European languages from other language families is ablaut, or a vowel change in verbs that signals different tense and aspects: wg-sang-swig^fril-fetfyalbn^fty-jiew-jlown. Through the last ten centuries, about half of the over 300 originally strong verbs have disappeared. Over 80 originally strong O£ irregular ablaut verbs have become weak, so that they now follow the regular -td form of the past tense and perfect participle endings. But almost 70 strong verbs remain strong, and a handful of originally weak verbs, by a kind of reverse analogy, have joined them. Excepting be and do, strong verbs in English have at the most three forms: infinitive: sing, walk, hurl; past or preterit: sang, walked, hurt; and perfect participle: sung, walked, hurt. Some OE verbs, on the other hand, had four, (1) The infinitive form served as the root, for the present form: beodtm-ic beode (I command). (2) A second form served as the form for the past singular first and third person: kfhe bead (I/he commanded). (3) A third form served for past plurals and past second person singular; puiwejgei/ik budon (you/we/you/they commanded). (4) A fourLh form served for the perfect participle: reftsebbe(ge)boden. (I have commanded). Ge~ was a typical prefix in the perfect form. 11.1: Traditionally, the OE strong verbs are organized into seven classes, depending on the kind of ablaut in the vowel. For ModE strong verbs, however, such categories are largely useless for three reasons: because sound changes in the last ten centuries have obscured the original basis for classifying them, because a fairly large number of weak verbs have become irregular through other sound changes, and because inconsistent analogy has made speakers choose ипpredictably from among the four OE forms to select the three forms we now use. Originally strong were the OE ancestors of all the ModE verbs listed below. First is an example of the class that is still strong: for Class I, rise-rose-risen. Then follows a list of other* that are still irregular verbs: write, ride, and so on. Following that in parentheses are ModE verbs that were once strong in that class but are now weak: Гп Class I arc abide, glide, and so on; in II arc reeked, shoved, chewed, and so on. On the PROBLEM

GRAMMATICAL CHANGE L THE VERB PHRASE

261

basis of the model and the other verbs, how would you guess a sampling of those weak verbs would be pronounced today if they were still strong? Class I: rise-rose-risen: write, ride, bestride, bite, shine, drive, slide, Strike, smite {abide, glide, spew, writhe). Class II: freeze-froze-frozem creep, fly, cleave, shoot, hse.fiee, choose {reek, shove, chew, rue, brew, lie [prevaricate], seethe). Class III: drink-drank-drunk: shrink, begin, sink, sing, run, swing, ring, Slink, spring, swim, bind, wind, spin, find, cling, grind, sting, swell, win, fight, burst (climb, burn, help, delve, melt, carve, starve, mourn, spurn, bark, yell, yield, swallow). Class IV: steaf-stole-stofen: break, beer, tear, come, shear. Class V; weape-wove-woven: tread, eat, give, see, sit, lie, speak {knead, weigh, fret, wreak, scrape, reap). Class VI: shake-shook-shaken; draw, forsake, stand, slay, swear, wake (fare, bake, wade, shave, wash, heave, step, gnaw, laugh, flay, engrave). Class VII: fall-fell-fallen; hang, hold, know, mow, sow, grow, hew, blow, weep, read, beat {wax [grow], flow, dread, row, blend, crow, leap, span). 11.2: Here is an artificial example of "illiterate, uneducated speech." Comment, / writ hint that the flowers ris because he holp me last year when we dim the hill in plant them. When we sot on the top, we seed all the stones, so we ,flang them in the pond. The big ones we digged a hole for and hove them in. (Note alhhc irregular forms like ris, holp, clim, and sot and so on have occurred in various dialects of British and American English.) (4, T74) PROBLEM

The kinds of judgments that prescriptive grammarians make about verb forms like these very often take on a tone of divine authority. The fact is, "correct" and "incorrect1" verb forms are more a matter of historical chance than grammatical inevitability. Many originally strong verbs have become weak. Those speakers who say He dim the stairs happen to be using the historically attested to form. Some weak verbs, on the other hand, have become strong: He dug the hole andfiangit hi. It just so happens that dug has become an acceptable new past form. At one time, digged was the ''correct" preterit. Flang, on the other hand, is not, even though iL exactly models the preterit of sang, swam, rang, and others. And in many dialects, of course, some of the verbs which are strong in standard English have been regularized to become weak. Just as knead, weigh, wreak, and reap have been regularized, so have go to goed, draw to dravted, swing to swinged, see to seed, and so on. It just so happens that those who used new weak preterits like weighed and wreaked were apparently more socially prestigious than those who regularize and use seed, goed, anddrawed. (170)

262

QRAMMAH AND SOUND

PROBLEM 11.3: Are there any irregular verbs which are in the process of becoming regular today? How do you know? The model for regularizing strong verbs, of course, is the paradigm of "weak verbs, which included by far the greatest number of verbs in OE. With some except ions, weak verbs did not change their stressed vowels to signal preterits, but ra the r added a feature that d isti ngu is ties t he Genna trie languages: the dental preterit, the source of our -ed past tense ("dental" because the /d/-/t/ sounds are made by placing the tip of the tongue against the back of the upper teeth): in iufie (I love) vs. ic lufadt (I loved). In some cases the OE verbs added just -de: ic deme (judge) ic denude (Г judged); in others, -tde or -ode: ic trymede (I strengthened), re endode (I ended). PROBLEM 11.4: In OE, preterit endings with vowels—ode and -ede—were always pronounced as full syllables: la/ode was a three syllable word, for example. Today, it is a one syllable word: laved. At some time before ModE, the final -e was lost and the penultimate (second to last) vowel was syncopated, at dropped. Suggest a way to determine when the final -e was lost and when the vowel before the /d/ was lost, PROBLEM 11.5: Fs it possible to predict the exact pronunciation of the regular dental preterit endings in ModE? Consider these words: pushed, snapped, danced, bombed, culled, buzzed, hissed, ended, lurched., judged, fitted, rubbed, ripped, longed, routed, star Jed. How would you explain the inconsistency between spelling and pronunciation? (Among the irregular verbs now in ModE arc these: lead-led, speed-sped, deal-dealt, set-set, meet-met. But these were originally weak in OE, with no vowel alternation. The ModE variations resulted from a sound change that we shall look at more closely when we discuss phonology.)

Person Endings In addition to these vowel changes and the dental preterit to indicate tense and perfect forms, OE, ME, and Harly ModE had other endings that also indicated the person and the number of the subject. Some of these can still be found in ihe eonscrvative usage of the King James Bible (1611): Яви saith the Lord. . . , Kno\v$1 thou not this . . . , If thcu hadst. . ., When thou wast young.... By this time, however, it had generally disappeared from standard written English and probably in educated spoken English. In some rural areas of England, however, it has been reported as late as this century. If we compare the regular weak endings from OE through ModE, we can see better how they have changed. In the ML list in the table, the first

GRAMMATICAL CHANGE: THt VERB PHRASE

263

form represents the Northern endings; the second the Midland; the third, the Southern. lirst

Second

Third

Singular

OE ME

ic cysse pit cys(se)st icfi к iss(e)!kisse!k:sse pu kisses)к tsses(t)j

EModE

I kiss

you kiss

hecy.isep kissestkissep he kisses

Plural <sitim.1 for all ijursoiisj

OE ME EModE ModE

Preterit OE ME

we cyssafr wekissesjkisse{n)!kissei> we kiss we kiss

EModE ModE

ic cyste kit kistjkistejkiste I kissed I kissed

Plural OE ME EModE MoilL

we cyst Oft we kislikiste(n)lkiste(n') we kissed we kissed

pucystest pu kistikistestlk is test youjthoukissest you kissed

he cyste he kistjkisteikiste he kissed he kissed

IJ.6: Compare the endings in the table above. What endings completely disappeared? What endings in ML are not predictable from OE? What endings in ModE are not predictable from ME? How might they be explained? FROBLEM 11.7: Go through the iame sequence of phonetic changes described for noun endings: (1) Unstressed vowels level to /»/. (2) /an/ endings become /a/. (3) Final /э/ endings disappear. What contrasts are lost first? What are the exceptions to this sequence uf changes? PROBLEM 11.8: A number of explanations have been given for the adoption of the third person -s. Here arc five. Comment. 1. 1. 3.

The third person -s allowed poets more rhymes than -(e)fh endings. A regular sound change changed hnal /S/ as in wreath to /я/: kisseft > kisses. /O/ is a difhcult sound to pronounce in conjunction with other consonants and was simplified to /s/.

164

4. 5.

GRAMMAS AND SOUND

The -$ ending [s an analogical extension of the third person is. It was borrowed from Danish.

Two minor categories of verbs that do not conform to these patterns are the preterit-present verbs and a very few anomalies. The preterit-present resulted from verbs that had an original past tense vowel in the present form and Lhen added the dental preterit: agan (own-ought), синпап (can-could), magan (may-mi ght), * mot an (an archaic №»fe-must; the * means here that no actual citation has been found for the word), scuiun (shall-should). Wilt with its iifLernaLe wouldcome from a regular verb witlaa. The most anomalous verb in English, be, has in its paradigm the descendants of different words: beon, в и н , and *esa». It also varied markedly from dialect to dialect in its occurrence with various numbers and persons. In the ME examples, the slashes separate Northern, Midland, and Southern forms respectively. OE

ME

ModE

iceom,beo l?u eart, bist he is, Ы we simian, be-o

am, be/am/am, em ert, es, bes/art, best/art, best es, bes/is, beoj?, bu^/is, beof>, bej? ar(e)r es, bes/arn, Ъеор, ben/beof, hep

am are is arc

he wses we waeron

was, wes/was/was was, wes/wore, were, wast/ weore was, wes/was/was wer, war{e), wes/woren, were(n)/ weore, wsere

was were was were

In OE, there may have been some slight semantic differences between bear: verbs on the one hand, and eom, zsrt, is, and sindon on the other. The be verbs frequently expressed the tense of repeated events or invariable facts, which in ModE might he rendered They be always ready or The door always be open al ten o'clock. The e&mjeartfis verbs indicated a condition whose permanence was not emphasized: He is here, lam happy. Among the о [her anomalous verbs were the ancestors of have, live, say, do, and go, whose pasi tense, wenJ, is derived from the pasi of wendan, the ancestor of the verb wend, as in " t o wend one's way home." PROBLEM 11.9: Why would one suspect have, sayt do, and go of being anomalous verbs in OE strictly from modern evidence? PROBLEM 11.10: What useful comparisons can be made between the changes in noun phrase and verb phrase inflections? Why do speakers uot confuse the ModL -s endings for plura!, genitive, and third person singular? What might be an internally motivated reason why the Northern tendency to add

GRAMMATICAL CHANGE: THE VERB PHRASE

263

-s endings to all present tense verbs did not lake hold when other Northern inflectional innovations did? (5, 150) The Progressive -ing, и Problem The form of the ModE -ing ending in (he progressive now happens to be identical with the -iag ending we use to create gerunds: He was singing (progressive) vs. His singing was awful (gerund). This nominalizing or nounmaking -ing goes back to the OE suffix -ивц{е). The original O£ progressive ending was -ende: He wxs ssngende. By MR, it had become -and(e) in the North, -anti, -ends, or -ing(e) in the Midlands, and -inde or -mg(e) in the South. How -ende cither became or was replaced by -ing is not at ali clear. It has, in fact, been debated whether there really was a progressive in OEon the model of ModE be + У-тц. Patterns like bean +• V-ende occurred very infrequently in texts originally written in O t , and only a little more frequently in translations which attempted to reEider Latin periphrastic verb forms. Even in ME, the progressive was uncommon. Only in the eighteenth century do we find it occurring with the passive. // wax being built (replacing the older It was building or h was a-building). And not until the nineteenth century do we find it occurring with hace or be: He is having a party—He is being careful, (ЛИ these new passive usages, incidentally, usages which are entirely correct in the twentieth century, were roundly condemned by the grammarians of the nineteenth century.) Some have suggested thai during ME, certain grammatical patterns of the gerund form with -ing and the progressive form with -ende were so similar that they merged: Thei hud ben afyghtUtg -Thel had been fyghtende. The a is a reduced form of an or at which would have as its object the gerund: Thei had ben onfyghilttg. The usage is still common today in rural areas: He's gone a-hunfm\ From the same source comes aboard (on hoard), mvay (on way), aside (on side), and so on. The gerund form may have come to be reinterpreted as a progressive when the u weakened and disappeared. The relatively frequent occurrence of this gerund pattern and the relatively infrequent incidence of the participle -ittgl-etiik gives some supporting evidence to this position. On the other hand, enough cases at-ende progressives dated before the V-irtg constructions have been found to suggest that the relationship may be a direct one, perhaps resulting as much from phonological confusion as from grammatical substitution. (22, У7, 151, 221, 230) PROBLEM 11.11: The -big has not yet spread to normal use with all verbs or to all instances of be. Try to formulate some general restrictions that would account for these next sentences: 1. 2.

* He was weighing I SO lbs.—He was weighing the chicken. * He was resembling his father.—He was imitating his fiat her.

GRAMMAR AND SOU МП

266

3. Л. 5, 6, 7,

*JIe is having a car.—He Is having a parly. *He is being fat.— He is being careful. *He is being old,—He is being helpful */ am liking is very much.—lam studying it very hard. •/ <m resettling that.—I am rejecting (hat.

PROBLEM 11.12: Listen closely to the pronunciation of words in ordinary conversations spelled singing, jumping, dancing, and so on. You can often hear something we call "dropping the -g" He was singin\ dancirC 'n around. Is this difference predictable or any grounds'/ (56)

The Imperative and SMEIjmu-ihi-

Finally, we might note that identifiable imperative and subjunctive verb forms have entirely disappeared. In formal ModE, we may use the same form as the past (were, in the case of be) to indicate counter-factual or conditional s t a t e m e n t s : If he came, say if he eomes,

we will...

we wauld.. ,1/1

. , Were

! he, I would....

was him, I vould....

But w e a l s o

We n o longer have any

distinctive imperative form, but now use the infinitive form instead: Slop thai f Be good.'

PROBLEM 11.13: If we reduce noun phrases and verb phrases to componentiai elements such as these below, how would a transformational grammar insure that the form of a verb in OF, MR, and early ModF. would be "spelled" in the correct form? (Review Problem 10.10.) Assume present lensc. What is relevant to verb agreement? (1)

demt judge) dem(judge) PROBLEM 11.14: Sketch a new system of verb agreement in which gender would be relevant to verb inflections. What other kinds of possible semantic

GRAMMATICAL CHANGE: THE VERB PHKASt

267

and grammatical distinctions among verb ending can you imagine? For example, could the age of the subject or objctt be reflected ? Hi* social status ? Could we inflect for lies? For irony? For probability? For judgment (i.e., fortunately or unfortunately)? For emphasis? For the shape of the object? For the value of the object ?

THE GRAMMAR OK THE VERB PHRASE

Now that we have outlined the forma! changes in the inflections from OE to ModE, we can examine how they function in the context of their verb phrase, and how a grammar that generated OE forms would differ from a grammar that generated ModE forms.

Auxiliary Verbs To explain how tense, -ing, and the perfect endings were distributed in OF, and ME and how the ModE patterns emerged, we must look at the part of the verb phrase that includes the auxiliary verbs: the modafs: ran, could, may, might, shall, should, will, wetuld, muss, ought; the perfect km^e: have gone; the progressive be: in going; and the passive be: was found, and the combinationi of these. In ModE, this structure allows several combinations: L 1. 3. 4, 5. 6. 7. 8. 9. 10. П. 12. 13. 14. 15.

He He He He He

He He He The The The The The The The

may has is may may may car car will car car car will car car should

have be has been have been was be has been being is have been has been being have been beinii

fixed the ear. fix the car. fixed the car. fixing the ear. fixed the car. fixing the car. fixing the car. fixing the car. fixed by him. fixed by him. fixed by him. fixed by him. fixed by him. fixed by him. fixed bv him.

The modal, the have, the progressive be and the passive be combine with the main verb into fifteen different sequences, all of which have been given a

26S

GRAMMAR AND SOUND

many complicated names by traditional grammarians: past perfect progressive, future perfect progressive, and so on. BuL just listing and naming them says little about their underlying structure, about the grammar of these phrases. If we note a few patterns, however, we can make some signiiicant generalizations: 1,

These elements may otcur in any combination with one another, but always in the same order: modal (may, will, shall, can, must)—have I had—

2.

The tense (present/past: will-would, have-had, is-was, and so on) is always indicated in the first verb of the phrase. The form of any individual verb—infinitive (go), progressive (going)., perfect (gone), or inflected for tense (goes, went)—always depends on the verb (or lack thereof) immediately preceding it. Tin at is, a. The verb following a modal is always in the infinitive or uninflected form (see 2, 5, 6, 8, 10, 13, 15). b. The verb following have к always marked [perfect] (see 3, 5, 7, 8, 11, 13, 14, 15). c. The verb following the progressive be is always marked with an -ing form (see 4, 6, 7,8, 12, 14, 15). i. The verb following the passive be is always marked (perfect] (see 915). If no auxiliary verb is present, then the main verb is inflected for tense, as in (1).

J.

We can now sketch some additional grammatical rules to explain these ModE. patterns. We shall renumber the rules omitting all rules irrelevant to the question.

1. S-^NP VP 2. VP->.AUX MV 3. AUX -* TENSE (MODAL) (HAVE) (ЪЕггвятЛп) 4. TFNSE^[±present]

BEptt==1vr)

/(№) 5.

MV-vV

API

The parentheses around MODAL, HAVE, B E ^ ^ , , , , and B E ^ , * , in rule (3) mean that any or none of them can be chosen in any combination, but that they must occur in their relative order. Next to be accounted for is the fact thai, the first verb in the phrase always carries tense. If no auxiliary verb is present, tense goes after the main verb: He goes. If tense were listed after each of these verbs as an option to

GRAMMATICAL CHANGE! THE VERB PHRASE

269

bctiin with, including the main verb, we would need я very complex ru.lt to exclude Lcnsc from nil except Lhc correct verb. But the problem can be solved by writing a rule placing TENSE in one position and then using a transformation to transfer THNSE to the verb immediately following it. The rule would have this effect on these two examples: 1-

TENSE

V

[—present]

run

V

-*•

[-present] lan

2,

TENSE

MODAL

[—present]

will

V

> MODAL

go

V

[—present] would

go

Exactly the same kind of transformation will also create the correct form of the verbs after MODAL, HAVE, BE p f ( w r 4 i 3 i i v e j and BE № a s , T 4 . Except for MODAL, each of these auxiliary verbs must also have associated with it an inflectional feature that shifts to the following verb: [perfect] from HAVE to whatever follows HAVE; [ing] from BE p j 4 1 s l . M 4 l 9 t to whatever follows it; and [perfect] from BE p t t = a J v i to whatever follows it. In fact, we can write a composite rule to include aJl these elements: [affix]

verb -»• verb [affix]

(Affix = any TENSE, [perfect], or [ing]; verb = any MODAL, HAVE, BE, or any main verb.) The rule simply states that whatever element is indicated by [affix] should be re-attached \o mark the verbal element immediately following. Therefore, we must expand the base rule for auxiliaries once again. BASE RULE: AUX-*TENSE

(MODAL)(IIAVE [perfect]) (BE p r a E . [-mgJXBEpiM.

TRANSFORMATION: [affix] V

[perfect])

v

ffi [affix] where [affix] = TENSE, [perfect], or [ing]. v = MODAL, HAVE, BE or any main verb V, PROBLEM 11.15: Generate four or five verb phrases, applying the transformation and spelling rules as appropriate.

279

OR АММАН AND SOUND

PROBLEM 11.16: Very briefly suggest bow a traditionalist or a deseriptivist would describe these patterns. One furtber question must be accounted for before we can explore the history of some of these patterns. As Rule 3 is formulated, it allows the passive BE to occur before intransitive verbs, generating ungrammatical sequences like *He was disappeared. (1) The passive BE must occur only with transitive verbs, and (2) the passive, since к is related to roughly synonymous active sentences: Tom saw Bill—Bid wan seen by Tom, must somehow be related to the same deep structure the active is based on. Some transformationalists have dealt with the problem like this: The В Е д а я . „ can be included in the AUX rule: A U X ^ T N S (MODAL) (HAVE) {Bli BI4iertlMl , e ) (BE p a i f l l v s ). But B E w s 1 V B is limited to occurring before transitive verbs: BE^,,,™: . V NP (where V is not be, become, a group of other verbs like weigh, resemble, behave, and some others.) If BE P 9 J l i V E is selected from the choices in the AUX, it automatical!}1 signals the need for a passive transformation: NP t -+NP2

X X

ВЕ № Н 1 И 1 BE,eil«

Tom -+ВШ

TENSE TENSE

[perfeCL] (perfect]

BEPMftm ВЕдавд,„

V V

N P ; by by

[perfect] sec Bill [perfect 1 see

Д NP, by Tom

The other transformation then shifts the [perfect] inflection feature associated with B E t a 3 a m from it to the following verb, giving Bill was see-n by Tom PROBLEM 11.17: Here arc some OE and ME sentences that represent the possible auxiliary verb combinations during those periods. Forms that do not occur in these lists should be considered ungrammaLical. For example, a form corresponding to he was being laught would have been impossible in OE fat least such a structure has never been found). How do these combinations differ from period to period ? Try to sketch some rules that might generate each set in their period. Note that in OG, the verb following a modal has an -an infinitive ending. Inflections are in boldface, OE 1. 2. 3. 4. 5.

He sceal dead sweltan. (He shall death suffer,) He hafafi hit onfunden. (Me has it found.) God bid eardigertde on heofonum. (God is dwelling in heavens,} к sceal bidentte been, (I shall waiting be.) He sceal nesan Ismahei haten, (He shall be Ishmael called.)

GRAMMATICAL CHANGE ! THE VERB PHF ASt

b. 7.

27i

Se boda h gefaren. (The messenger is departed.) lie vma gefongen. (He was captured.)

ME 8. 9. 10. 11. 12. 13. J4. IS.

He hath promyst it. / кап not ben well an we. (1 can not be well at ease.) Il will not be worshiptWe han been waitynge al this fourlermyght. Ich mihte habbe bet i-don. (I might have better done.) i she! be chciuuged. He hath been у-foundsThere was com another knyghl.

The most obvious difference between OE and ME on the one hand and ModE on the other is the relatively limited number of possible auxiliary verb combinations in OE and ME. OE allows only six possibie combinations; ME, at least Early ME, allowed eight: (Old English Inflections, -infinitive AN, progressive -ENDE, perfect -EN, in

1. 2. 3. 4. 5.

MODAL MODAL MODAL

BE B l l B .-AN BE e fhaf-1

6.

verb-AN verb-ENDL verb-FN verb-ENDE vcrb-EN

(as in (1) above) (as in (4) above) (as in (5) above) (as in (3) above) (as in (7) above)

verb-EN

(as in (6) and (2) above)

be-

These combinations can be derived from this rule and the same kind of affix-verb transformation rule described above.

TENSE

(MODAL

-[ HAVE

[perfect]/j [perfect]

J

What this rule stipulates is the limited combination of modals with passive and progressive BE's, and the fact that the perfect corresponding to ModE have occurs without any other auxiliary. This rule must be developed further, however, because the HAVE, an abstract symbol, is manifested in two ways, depending on the kind of verb it occurs with. With transitive and intransitive

272

GRAMMAR AND

verbs of motion, HAVE is spelled isaf-, the ancestor of ModE have. But before other intransitive verbs, the HAVE is spelled be-, wes-, or weard-, (The full infinitive form, beon, is the ancestor of our be; the full form wesati is the ancestor of wasjwere. Weorftan has disappeared entirely.) ME has all of these OE verb forms, but also allows three more: 7. 8. 9.

MODAL

HAVE HAVE HAVE

E^.-EN B E ^ -LN

verb-EN verb-ING verb-EN

[as in (12) above) (as in (II) above) (as in (14) above)

These new ME possibilities can be represented with this r u b ;

A U X ^ TENSE (MODAL) (HAVE [perfect]) ({op 0 ""'

["**£' .})

PROBLEM 11,18: What inflectional changes are involved here?

Tf we compare the rules, we can see that the perfect HAVE has been taken out of the disjunctive choice with the other auxiliaries and put into a linear order after the MODAL and before the ВЕ11Г1И and B E ^ , A further change in this HAVE, however, is that it is now manifested as a form of be before intransiLive verbs of motion, as have elsewhere. This rule, however, is too powerful for Early ME. Tt generates sentences with three auxiliary verbs like He moyhoie been fowtd and He may hare been singing, neither of which appears to have been possible for that period. In other words, the Early ME rule would have to have a further constraint to the effect that only two out of the three auxiliary elements may be chosen in any one phrase. But that is a very "messy" kind of rule. It does not have the neatness of the OE or later ME rule. Indeed, it may be the very "messiness" that led to the grammatical change that finally allowed a sequence of three consecutive auxiliary verbs. If speakers tend to formulate the simplest set of rules, a constraint on the number of verbs in a sequence would seem to be a kind of e>Lcrescense that would disappear. Another characteristic of ME distinct from OE is not represented in this rule. It was the emergence of do as a kind of all-purpose function word. In OL, do functioned as a main verb on the model of He did the work, and as an anaphoric or substitute verb as in He works and I do too. Probably out of the main verb do arose a causative-do pattern. Thi soul-cmtl"ich wile doringe (Thy soul-knell 1 will do ring; i.e., I will cause someone to ring your soul knell,), Hedede meyk a newe if one (he did make a new throne; i.e., he caused someone to make a new throne), Perhaps out of this causative-ifo, that often indicated the completion of action, arose what we now find in modern nonstandard spoken English, particularly in the American South and in parts of Newfoundland: J've donefinished,or more usually, / done finished.

GBA,MMAT:CAL CHANGK: THE VERB PHHASE

273

Yet another do-form developed later in ME, perhaps out of the causative, as its semantic sense weakened. This was the semantical^ empty do Lhat in later Mli and Early ModE poetry seems only Lo flesh out a rhythm of in prose, was a seemingly meaningless unstressed syllable: They did find the answer . . . , We did feme when. . . , Out of this do, probably, developed the interrogative and negative do's: . .. why do ye wepe, they did not weep. The empty do reached the height of its use in the sixteenth century when almost any sentence, verse or prose, might con La in one. The last examples of the causative do date from the early seventeenth century. The interrogative-da was developing as early as Chaucer, but the negative-t/w was a later development. In order to include the aspectual do as in He has done left and the empty do as in He did leave, we have to expand the rules for ME auxiliaries: TFNSE (HAVE

(MODAL) [perfect]

(DO

DO

[perfect]))

f-№"*BE

[РегГ^]\\

\ L W. Hug] J /

This will allow He may have done been gone from do 4- EN. and from the oLhcr do, He does go and He may do go, forms which have been cited. The causative do described earlier is not an auxiliary do, but more like the verbs tell, want, promise, and so on, Lhat take complements of the form / told him to leave, / want him to leave, /promise to leave, I Ms a main-verb do: I did [someone] make the house, parallel in meaning to / had him promoted, but lacking the noun object. Between ME and ModE, five important changes occurred. First, the causative-fin disappeared entirely. Second, the aspectual have + done + V-EN (he has done gone) form, a form frequently found in northern British dialect areas, became stigmatized as nonstandard. It was probably brought to this continent by Ulstcrmen, many of whom became slave overseers and probably passed the form on to Blacks. Third, the unstressed empty do in sentences like He did leave disappeared (or merged with emphatic do: He did letjve.). Fourth, as we shall see in more detail, this empty do may have become the source for questions with dat hrst appearing in ME: Did he go?, then later LO negatives: He did not go. The earlier forms. Went he? and He went not, occurred side-by-sidc with the newer forms, often reflecting social distinctions, but eventually died out except in the most formal public prose: Ask not what your country can do for you . . . , Beuts there a heart in this great land which. ., ? Finally, the last disjunctive choice in the auxiliary rule, the choice between BE M 3 S l 7 e and Bh p r D e r B a b l V 4 1 changed to a linear choice of either or both:

274

GRAMMAR AND SOUND

Be is watching, He Is watched, He is being watched. Before this time, before later ModE, the progressive sense of the passive was communicated in a sentence like The food was bringing in or The house wasa-buitding. As we have seen, the origin of this construction was probably the gerund: The house was on building. As (he on weakened to a-: The house was a-buiiding, and then disappeared: The home was building, a form identical to the progressive appeared, perhaps even replaced the -etide form. But in the eighteenth century, grammarians began to object to this construction, partly because it was bomonymous with the non-passive progressive: The kemsejwirat was building, and partly because they felt it not to be in accord with the "genius" of the language. One of the earliest citations of the modem finite passive progressive is from 1795: , . , like a fellow whose uppermost grinder h being torn out by the roots.. . . Centuries before this time, though, what appear to be participial modifiers with progressive passives had occurred: , . . he should mind in fear of fire, being burnt i'th hand for steeling ofsheepe. The potential confusion of the two forms, the earlier progressive passive {The house was building.) and the progressive active (The wind was building.) plus the earlier occurrence of this passive participial phrase led to the progressive passive as we now know it. Another pattern that flourished in northern British dialect areas (and still does) and which was probably brought to America by Ulster Scots are sentences with multiple modals: lie muy can go, /might couldleave. She will can stay. Now, however, like the He has done gone, also of Northern British origin, the multiple modal к stigmatized as uonstandaid English. (22, 32, 47, 49, 15J, 195,220,230)

Nonstariilard English

The fact that there are these similarities between Early ModE auxiliary structures and structures that characterize what has come to be known as Black English strongly suggests that the one has been a strong influence on the other. Had we the space to explore some of the finer details of early ModE, we would discover other similarities. There are some significant differences, however. One dialect of Black English seems to lack a form of the inflected be in sentences such as these; L 2. 3. 4. 5.

He in the house. She my friend. You a bad man, They playing do wnstairs. You covered with dirt.

GRAMMATICAL CHANGE: THE VERB PHRASE

275

At the same time, this dialect also seems to use the uninilcetcd form of be " i n c o r r e c t l y " in sentences like these: 6. 7. 8. 9. TO.

They be there tomorrow. / he ready if I cotdd, Whenever they be there, they be happy. She be take» care of They always be joking.

And some sentences appear Lo be like passives, but are not: 11. He been left, (i.e., he has left.) 12. She been worked, (i.e., she has worked.) 13. They been et. (i.e., they have eaten.) Some have argued that these differences in surface structures reflect deep structures in Black English that are significantly different from the deep structures of standard English, and that therefore, despite the superficial similarities between the two, Black and standard English are really two different languages, with all that that implies for education and social understanding. Indeed, the source of these differences has occasionally been attributed to the grammatical structures African slaves brought with them in the earliest days of slavery. Actually, a rather simple deletion rule seems to explain many of these patterns. Whenever Standard English speakers contract an auxiliary verb or the main verb be, the dialect of nomtaodaid English illustrated by these sentences will predictably delete the verb: Full form:

I am here. V

Contracted form:

I'm hers.

Deleted form:

I here.

V

I have gone. V Fvegone. V

Igotte.

I will be ready. V

ГНbe ready. V

I be ready.

These deletions are not invariable, any more than contractions are invariable in standard spoken English. BuL statistically, the contracted verbs do tend to be dropped, just as statistically, other speakers tend to contract them. The statistical trend can be charted across social classes and styles. As a speaker becomes more casual, he will, depending on his linguistic background, contract or delete more and more frequently. In fact, the very ability of a speaker to adapt his speech to social situations requires us to conceive of a grammar as an intersection between rules of the kind we have been describing and their statistically predictable application. Some grammatical rules have 100 percent predictability: The relative

276

GRAMMAR AND SOUND

pronoun will always occur as the first noun phrase in a full relative clause: the man whom I saw vs. * the man I saw whom. Other rules depend on variables related to social class, social situation, and speaker-audience relationship: the man wkatn I saw vs. the man who i saw vs, the man that I saw vs. the man I saw. But some other nonstandard sentence types resist this kind of explanation. One is I be here everyday, so /saw what happened. This does not. seem to be an ellipsized version of Г И be here everyday so.... In fact, this invariant be appears to be a characteristic of certain nonstandard North Лmerican dialects, especially, but not exclusively, Black dialects. It appears that this invariant be communicates a continuous habitual sense of time. Where standard English communicates duration through time with the usual inflected form of a verb or with a verb and adverb: He is always there, some dialects of English seem to signal the same notion of aspect with the form of be: Thus I be there means I am regularly and predictably there. Where this invariant be came from, however, is a bit of a puzzle. There was (and is) iit the southern British dialect of ModE a be in sentences such as these: I think he be a traitor. Where be the men ? We be ready to leave. If it be as you say, then let us end it. But these are simply the forms of the present tense of be used in Southern British English instead of is or are, often with plural subjects or subjunctives. By the sixteenth ccnLury, this particular form of be was more often found in the speech of nonprestige speakers. But none of these examples communicates the same habitual sense of time that the present nonstandard Fnglish be can communicate. It may be that the durative be has its origins in the OL beon, the /w-verb that often indicated duration, that after being used in that durative sense with some frequency in О Li, iL submerged into a lower class or lower middle class social dialect, was brought to this continent, and is now being observed for the first time. Support for this view and against the view thaL this durative meaning was brought with the slaves from their native African languages and invested in the be form is the fact that this invariant durative be has also turned up in Eastern Canada as well as in the American South, from whence it was apparently brought to the inner ciLies of the North, It is difficult to argue convincingly that Newfoundland Canadian whites could have adopted it from the relatively few Blacks that settled there or that exactly the same meaning could have independently evolved in exactly the same form in the two diafects. An anomaly among these verb forms, however, seems to occur in sentences such as He left, ain't he?T\\s full and < l correct" form would be He has left, hasn't he? The contraction and deletion yield the first part of the sentence: He has feft > he's left > he left. But the ain't he tag question seems to contradict the generalization that such tags normally repeat the subject and auxiliary verb and add a negative if the main clause is affirmative and vice versa. In He left, ain't he? ain't appears to be a contraction of am no(. But

GRAMMATICAL CHANGE: THE VERB PHRASE

277

that would imply the full form of he left, ain't he? was cither *)te is left, isn't he? or *he has left, isn't tie? both of which are impossible in English. Historically, ain't is a contraction of am not: am not > amn't > aarft > ain't. If it is now ** ungrammatical" in some upper middle class dialects in our north centra! states, it was as late as the turn of the century freely used by many upper middle class educated speakers in the southern part of England and is used today in the casual speech of a good many highly educated southerners and westerners. To call it totally ungrammatical is to betray one's geographical biases. That it is now considered such is a sociological, historical accident. But ain't has a second source. It also descends From a contraction of have not. Such forms as ain't in He's got the money, ain't he? evolved from hain't to han't to hain't to ain't, making ain't a contraction of either have + not, has + not, or am + net. The occurrence of sentences like He dead, ain't he? illustrate the common tendency to generalize a form. Although ain't historically is derived from the contraction of am not, it is used as the tag for sentences with other forms of be. This can be demonstrated by sentences suc-h as He ain't dead, is he? Thus what at first glance may appear to be an anomaly—ain't occurring where we would expect haven't or em not—is, if not sociolinguistically prestigious among some speakers, historically consistent and as logical, indeed more logical perhaps than 1 am here, aren't П, where the aren't suggests a plural subject *are I? Another grammatical form that has been uniquely attributed to Black % English dialects is the use of been in intransitive sentences of the form: He been et (et = ate). This is not a passive form, but rather a form whose meaning is probably the completion of an action in the remote past, as opposed to a form like He done ate, which represents the completion of an action in the less remote past. While there appear to be no citations of such verb phrases in any extant English texts from older periods, it too has beer found in the speech of Newfoundlanders. (39, 40, 55, 54, 77, \ 19, 135, 197, 211, 212, 221, 224) Whatever the origin of the differences between standard and Black nonstandard speech (though the weight of the arguments is on the side of British rather than African or pidgin influence), the reason for the difference today is social isolation, F.xample after example of preserved archaic forms can be found among groups of speakers who have been isolated from a changing standard, isolated because they have been cither geographically or socially cut off from the influence of those who set the standard. Those who are geographically isolated have little opportunity to hear the standard model. Those who are socially isolated often have no expectation of moving up and out of their social class and so do not seem to imitate the speech of those in the more prestigious groups as strongly as do those who feel themselves to be upwardly mobile. (1,200)

27$

GRAMMAR AND SOUND

TRANSFORMING THE VtRB PHKASt Changes in the auxiliary verb system since ОБ have been accompanied by &ome other changes in [he verb phrase. They involve the way auxiliary verbs operate in questions and negatives, the way some verbs and prepositional phrases have restructured themselves into two-word verbs and direct objects, and the way one pattern in OE and ME has disappeared entirely.

Negation

Closely tied to the patterns of the auxiliary phrase are negatives. Because this is such a complex subject, we shall deal with only the major patterns and those in no great detail. ModE negatives may occur in these verb phrase patterns: 1. 2.

1 will not tease, / have not left,

3. 4.

1 did not leave. I was not leaving/happyj in ?f)e fiouse.

The negative occurs after the first auxiliary verb {MODAL, have) or after any form of be. If no auxiliary verb is present, then a do is needed as in (3). Negatives can also occur in these palLcms: 5. 6. 7. 8. 9.

Noflowers bloomed (compare: Flowers did not bloom,) The man had no money (compare: The man did not have any money.) The man never arrived (compare: The man did not ever arrive.) He wen! nowhere (compare: He did not go anywhere.) J saw nothing (compare: L did noL sec anything.)

Because the negative patterns that negate a verb: The man did not have any money are often synonymous with negatives that negate a noun phrase: The man had no money, negation seems to be a kind of meaning that ranges over a whole sentence, an element that can be manifested in different places while the sentence it occurs in retains the same meaning. In other words, different surface patterns can represent the same deep structure, the same underlying semantic structure. One way to deal with the problem is to generate the NEGfative) element separate from the rcsL of the sentence and then shift it to the appropriate place in the sentence: S ->• (NEG) NP VF. Since the not may occur after an

UKAMMATICAL CHANGE: THE VERB PHRASE

279

auxiliary verb or be but before the main verb, we need a transformation of this sort: NEG -^

X TENSE X TENSL

(y) (v) NEG

(where v = MODAL, BE, or HAVE )

The NLG is shifted after Tense, and if a modal, be, or have is present, then after it too: NEG -±

I TENSE be I TENSE be NEG

happy happy

Once this transformation has occurred, the TENSE is then attached to the immediately following verb. But when no modal, have or be is present, then the NEG is moved behind the TENSE, but it will not be after the main verb: NEG -*

he TENSE he TENSE

NEG

go go

In this case, the TENSE cannot move past the NEG, so we insert a do to "carry" the tense: ,

he TENSE NEG he do-TENSE NEG or: he docs not

go go go

(assuming TENSE is [ +present])

Negatives that attach to noun phrases are more complicated to describe because they are closely associated with indefinite pronouns: anyone-no one; sottiebod^anybody-nobody ; .wrtiew/iere-at and because of a some-any variation (in standard English): I have some money-i don't have any money-I have no money. Without getting into the complexities of the problem, we wjlj only note that the NEG, if it is not moved after the TENSE in the ajxiliary phrase, may be attached to an underlying indefinite ART in a noun phrase: NEG -* or:

I TENSE have I TENSE have I

have

ARTICLE NEG

money money

no

money

This is a grossly oversimplified, incomplete, and in many ways misleading description, but it will serve our purposes here. The most important point to

280

(JKAMMAR AND SOUND

note is that in standard ModE, only one occurrence of NEG in a sentence is allowed (barring prefixes and sulfixes): */ don't hat» no money is in standard Modi- ungrammatical. PROBLEM 11.19: Here are some examples of negatives from ОЁ, ME, and Early ModE, Translutions retaining Lhe negative pattern follow in parentheses. What kinds of rules would account far the differences between the various stages of the language ? (There are other patterns in addition to these,) Old English I. 1. 3. 4, 5, 6. 7, 8.

Ne mihie he bidden. (Ne might he pray.) Ne fuafon g? naht besorgian. (Ne need ye not worry.) Ne demdon we. (Ne judge we.) Nxbbe ic rttcitlg sa'pu. (Ne-have J ne-any ships.) Na>s he snotor. (Ne-was he wise.) Ne ic ne herige ne ic ne ule. (Ne I nc praise ne I ne blame.) Ne hit nxfre ne gewwSe. {Ne ч nevtr ne happen,) Nis nsnis ЯИ» bitSe. (Nc-is ne-any so happy)

Early Middle English 9. 10. 11. 12. 13. 14. 15.

Ic ne seye not. (I ne say not.) He ne held it nogkt, (He ne held it n o t ) fie ttoSde slepe ttl noon house. (He ne-wonld sleep in no house.) There па? па mutt nowhere so i-ertuous. .\'e tnketh nothing So hold of no men ne of no womman. He ne speketh new. No топ nute don hym no good. (No man ne-will do him no good,)

Later Middle English 16. 17. 18. 19. 20. 21. 22. 23. 24.

/ stye not lhe wordes, He held it aoght. We save them navtt. We sawe nawt the knygfttes. A man has not all his wylL He wolde not came inn. Thei may nawi singe. / am not coUl and nakyd. // Is not ionges gon. (It is not long ago.)

We can see two very significant differences among OE, ME, and ModE. First, multiple negation was once common in English. The change requiring that only one NEG appc;ir in a sentence is a relatively recent rule in the dialect of Early ModE that has become the sLandard, In fact, il is one of the

GRAMMATICAL CHANGE: THE VTRB PHRASt

281

few instances where prescriplive grammarians may have encouraged a tendency already at work. Standard Early ModE had already begun to favor single negation before the preseripiion against multiple negatives appeared in usage books of the eighteenth century. When prescriptive gram mars Jike Lowth's extremely influential Short Introduction to English Grammar stated that , - - two Negatives in English destroy one another, or are equivalent so an qfiirmative, the tendency was simply defined us the logieally prescriptive norm For standard speakers. His advice has been repeated so often aitd so strorgly that few English speakers who have endured our educational system can fail to recognize that No one didn't haie no money is not a favored construction among educated speakers, despite the weight of respectable English history behind it and the testimony of numerous other languages in which multiple negation is not factored out like an algebraic formula, languages in which the more negatives there are in a sentence, the more negative il is. The specific change we would need in a rule to generate multiple negation in OE and ME is one That would "copy" NEG in several specific places rather than one that allows only one placement of NEG. The second major change involves the loss of one option in the placement of negatives after verbs. As late as the end of the last century, the i/fl-negative and the negative without do were not unknown; [ did not leave-I left not Presently, though, the do- less negative has largely disappeared from standard informal educated English, (53, 94,97, 132,221) ^PROBLEM 11.20: What are we to do with the following sentences? la. lb. 2a, 2b. 3a. 3b, 4a. 4b. 5a. 5b. 6a, 6b.

/ haven't any money. / dont hope any money, You needn't stay. You don't need to stay. He dont never be there. fie ain't there now, Can't nobody tell him what to do? *Cant George teit him what Го do? A in't no way he gonna find out. * Ain't that way he gonna find out. Didn't nobody tell me that? * Oidni that man tell me that?

Why are 4b-6b wrong? What does the fact that you know they are wrong tell you about the grammatically of nonstandard dialects? Questions

Like negatives, questions in Mod L also relate in some important ways to the grammar of auxiliary verbs. English has two kinds of questions: Those

2S2

GRAMMAR «ND SOUND

we a n s w e r with a yes от no, a n d t h o s e w e a n s w e r with i o f o m a t i o n : ( I ) Can he come (hen? When can he come ? (2) Has he done it? How has he done it? {}} Was he going there ? Where wax he going ? (4) Did he see that ? What did he see? The pattern* are similar. In yes-no questions the first element is always a modal, hate, ov be, or if none of these is present, then a form of do is required as in (4). In information questions, the first clement is the interrogative pronoun. Then if the pronoun asks for information about any element in the sentence except the subject, the next element must be a modal (J), have (2), be (3), or do (4). But if the interrogative questions the subject, then the main verb may follow: Who left? What happened? One way to explain question patterns within the framework of a transformational grammar assumes that all sentences, not just questions, begin with what is tailed a performative utterance having the force of I slate, 1 ask, I order, and so on. Among the reasons for assuming this is the problem involved in a sentence like Frank!}', I hare no time or Frankly,did you brush your teeth ?The adverb describes the attitude of the speaker: ! say frankly от J ask frank h\ . . . We have to assume in such sentences an implied subject-verb which is deleted. There is a good deal more evidence, but it need not concern us here. We can represent the statement, question, and imperative performatives with a shorthand ST, Q, and IMP in the first rule:

(NEG)

NP

VP

[IMPJ The deep structure of a question would then look something like this: Q

he TENSE

will

sec

something

A transformation then (1) changes the element questioned into an interrogative pronoun and (2) shifts it to the beginning of its clause: 0 he TENSE will see something -^Q he TENSF, will see what -+Qwhathe TENSE will see If the series of transformations stop here and the Q is realized as / ask, then we would have the sentence: / ask what he will see. But if we go on to delete the performative, the Q, the lask, then we are left with a sequence of elements, what he TENSE will see, needing one more transformation. In this case, the TENSE and the first auxiliary verb or be is

GRAMMATICAL, CHANCE: THE VERB PHRASE

283

moved from its position after the subject to a position directly after the interrogalive pronoun:

-*• •+ or:

Q what he TENSE will see what he TENSE will see wbaiTENSEwlBbE see? what will he see? (assuming TENSE is [+present!)

11.21: Perform the question transformation on these strings below.1 Assume that the indefinite pronoun, whatever it is, is to be transformed into the corresponding interrogative word—who, when, why, how, where, what. PROBLEM

1. 2. 3. 4. 5.

Q I Q he Q she Q they Q you

[—present] [+present] f-l- present] [—present] [—present]

be somewhere may ask someone have -en buy something be -ing work sometime shall say that for some reason

But if the first verb after the TENSE is not an auxiliary or form of be, but rather the main verb, then only the TENSE can be moved: what what

he TENSE he

mid find

TENSE * This leaves TENSE without anything to "carry" it, and we insert the do again, just as we did with negatives. what TENSE he find what do TENSE he find or; what did he find

[assuming TENSE is [-present])

11.22: The lack of a do after the interrogative in a sentence like Who saw the boy? seems to be an anomaly in the pattern. Go through exactly the same transformations described above: (1) Indefinite pronoun becomes a WH-word, (2) WH-word shifts to beginning of clause, after Q. (3) Q is deleted. (4) TENSE and, if one is present, the first auxiliary verb shifts to a position directly after the WH-word. (5) Attach TENSE Lo the following verb. (6) Jf some element intervenes between the TENSE and a verb, insert a do. Perform those operations on this string: Q

someone

[-present]

see the boy

1 A string is a sequence of symbols generated by base rules.

234

fiRAlWMAR

AHD SOUND

PROBLEM 11-23: If we assume thai questions of the WH-word form follow the pattern described above, what would be the parallel derivation for yes-no questions? Did you beat*? and Did you leave or not ? are semantical^ if not stylistically identical. Presumably, they are related to a common deep structure. How does this fact bear on the problem of generating,^-™ questions? PRonLFM 11.24: Here are some questions based on OE, ME, and Early MudE sentence patterns. How do they differ from Modt question patterns? How would the rules suggested for ModE have to be revised? OE 1. 2. 3. 4. 5. 6.

Fart }>u se сynirtg ? (Arc у о u t he к i ng 7) Hwxt sxgst fat ? (What say you ?) Hwilee fixes xcfehst }m ? (What fishes catch you 7) Hxfst pu siagt gefangen? (Have you any caught'.') is pes ofpinum zeferuni ? (Is this (one) of your friends 7) Ne canst pit huntian butoti mid netfum ? (Ne can you hunt except with

7. S. 9. 10.

nets.') Ни gefehsi pu fixes? (How catch younshes?) iiivst pu hafoc ? (Have you hawk'.'} Hwmt canst pu sdepptin ? (What tan you create 1) mile ge beon beswmgen on ieomunge ? (Will you be beaten in learning 7)

ME 11. 12. 13. 14. 15. 16. 17. 1*. 19. 20.

Hwat is pe light ? (What is the light ?) Gafve the maun anythyng? (Gave you the man anything?) /W^nor.^Heardyounol?) To worn stiufd he speke ? (To whom should he speak?) Is that svtke? (Is that truth?) Shat I telten wnrc ? (Shall 1 tell more?) Has he a long srtewte ? (Has he a long snout ?) IWojAii/*uwWi'(Whosha]|hLinLhere?) Is youre chytda knave? (Is your child a knave (boy)?) How mow pey /wn shryie par synne? (How may they then shrive that sin?)

Karly ModE 21. 22. 23. 1A. 25.

Ride you far today ? Is this your friend? Wilt you be there? When left he? What think you now?

GRAMMATICAL CHANGE: THE VERB PHRASE

285

Who goes there ? Do you hear, porter ? Shall we go draw our numbers ? How many hast thou killed today? Why dost thou bend thine eyes upon the earth ? What sayst thou to her ? What seem I thai I am not? Who did strike out the light? PROBLEM 11.25: Compart the rule changes for negatives and questions. Why the rules changed in the way they did is difficult to determine. In the cast of negatives, it may have been that speakers felt negatives should always occur before the element that carried the main semantic content, being negated, the difference between He came *- not and He did not -*• come. We find negatives always preposed in noun phrases: no man, not a single person, for example, rather Elian *man no or *a single not person. We must also place negative adverbs of frequency before rather than after the verb: He hardly slept - *He slept hardly. He had Scarcely te/l when . . . - * He had left scarcely when, . . . On [he other hand, we can place non-negative frequency adverbs either before or after the verb: He often slept - He slept often. He usually left. He left, usually. In regard to questions, it may have been felt that the principal semantic ^erb should always occur directly after the subject, as it normally did when the first verb was an auxiliary: Larlv MIKJF: Mod t :

Will you SO? Hare you gune ? Are you going ? Went you? Will you go ? Have you gone ? A re you going ? Didyou go ?

PRODLLM 11.26: What might explain why speakers wanted the semantic verb to follow the subject? PROBLEM 11,27: What are we to do with the following sentences (Review Problem 11.20J? la. Ih, 2a. 2b. За. 3b.

Haiv you any money ? Dn you have any money ? Need I stay? Do I need to stay ? Во he always be there ? Is he there now ?

Whatever the internal and external causes that triggered the change, by the end of the eighteenth century, the negative and question had settled into

286

ПЕЛМНАН AND SOUND

relatively stable patterns. The negative now regularly occurs before the element being negated: So money, not enough, did not leace. The main verb in questions remains after its subject; Did he curae ? Both patterns were created simply by redefining what went into the v in the transformational rule.

Кesfrill-tunny the Verb Phras*

So far, we have examined several grammatical constructions that once quite different in OE or ME and have completely changed in ModE, But in some other cases, the change is one that is constantly affecting areas of the grammar. Many individual nouns, for example, are still being converted into verbs: The satellite is in orWt. -> The satellite is orbiting. And many verbs are still converting from one class to another: The satellite orbits, ~^They orbit the satellite. -> The satellite orbits around the moon, -> The satellite orbits the moon, -s- The moon was orbited by the satelliseThese arc changes involving the distribution of individual words. Other syntactic changes involve new transformations. These we have seen in new negative and interrogative patterns. More basic changes involve the deeper structures of a language, such as those we have seen in the rules for the auxiliaries. There we saw a simplification of rules stretching over 1СЮ0 years. A somewhat more abstract change involves the restructuring of a group of elements so that their sequence is not changed but their constituent structure is. For example, a sequence of elements like X Y Z could be restructured cither X (Y Z) or (X Y) Z : (2)

л

(3)

Each of these three structures is represented by the order X Y Z, but they have different structures underlying them. One example of a structural change like that from (2) to (3) is illustrated by sentences like lie talked over the problem. While in OE there were verb + particle constructions, many of those in ModE have evolved from verb + preposition constructions. Until about the fourteenth century, while there was still some degree of flexibility in where the preposition could occur relative to the noun-object, it. was possible to place the preposition (or particle) either before or after the noun. The earlier tradition of separable prefixes with verb* much as in Modern German, also contributed to this variable placement. As a consequence, some prepositions have become scmantically more and more closely associated with the preceding verb until they become a semantic

GRAMMATICAL CHANGE: THE VERB PHRASE

287

unit. In conjunction with the variable preposition placement, we developed a pattern in English allowing either 1 held up the man or 1 held ihe man up. Thus holdup is semantically a unit but still grammatically composed of two parts. The fact that we also require Me was holding up the man rather than * He was hold иррШ tfie ™an к further evidence of the semantic-grammatical ambivalence. The grammatical restructuring that went on—and is still going on with many other such pairs—can be represented like this: Earlier:

V

Later:

PKEP

KP

I

I

^^

look

over

the Ыю к к

look

PART

NP

over

the books

(We rename the PREPOSITION a PARTICLE because as part of the verb, it is no longer a preposition.) This sort of construction differs from VERB + PREPOSITIONAL PHRASE in some important ways, but much the same kind of restructuring is going tin with modern verb 4- prepositional phrase sequences like He looked m the man. VERB + PARTICLE constructions allow passives: Somebody held up the man -*• The man was held up by somebody. But VERB + PREPOSITIONAL PHRASE of the following sort does not: He lives in Cleveland-* 'Cfeceland is lived in by him. He arrived at noon, -э- *Noon was arrived at by him. He sings off key. -*• *Key is sung-offbyhim. But prepositional phrases after verbs of this next kind do a I low passives: He looked al the matt, ^y The matt was looked at by him. He asked for my friend. -> My friend was asked for by him. The beginnings of this passive with the preposition go back at least to Caxton and Malory in the laLc fifteenth century: / wyile that my moder be sen! for thai he is thus eotnplayend on. But it is a pattern that has become much more common in modern times. The change undoubtedly was encouraged by the semantic restructuring of VERB +• PARTICLE passives, so that now, virtually any VERB + PREPOSITIONAL PHRASE that eommunicatcs an action directed toward the object of the preposition seems to allow a passive, suggesting that the combination is becoming a semantic unit. Compare He held on Io the box and He stood beside the door. We can say The box was held onto but *The door was stood beside seems much less likely. Those VERB + PREPOSITIONAL PHRASE constructions that do not allow passives are ordinarily verbs followed by adverbial prepositional phrases of time, place, and manner—prepositional phrases in which the noun is not semantically "acted o n " by the verb. Compare He arrived at Cleveland

2SS

fiRAMMER

AND SOUND

and *Cluwland wot arrived at as opposed to He arrived at a conclusion and A conclusion was arrired at. One way to account for this grammatical difference is to assign sonic adverbial prepositional phrases—tbose not closely tied to the main verb—to a position in the tree by means of the base rule* we have been discussing that would represent the less close connection:

PP

NP

VP

N

t

P

he

arrived

at

/

f

\

NP noon

T(i distinguish this from He arrh-ed at conclusions, we have to associate at with the verb, but we must at the same time insure that at is still classed as a preposition. For like prepositions, it can be brought forward in questions and relative clauses: What time did he arrive at ? The time which he arrived at

Л f what time did he arrive 7 The time at which he arrived

What conclusions did b e . , The conclusions which h e . . .

At what conclusions did he ,, The conclusions at which he .

But the ai in arrive at conclusiOfiS is still not as free as a particle: *He arrived conclusions at. One possible solution is to write a new base rule that would generate a deep structure of roughly this form (compare above):

NP

GRAMMATICAL CHANGEJ THE VERB PHRAJE

289

We would be asserting that at two» is not part of a verb phrase but that at conclusions is part of the verb phrase, And it is that closeness to the verb that allows the passive transformalion. Along with The great increase in verb + panicle constructions, this is a change that is quite in keeping with the kind of general changes we have seen in English over and over again through the last thousand years: English is continually becoming a more periphrastic language. The auxiliary verb combinations have multiplied; prepositions have laken over the funelion of the dative case and the genitive case; articles have become obligatory for singular count nouns. Comparison with more has become more frequent than comparison with -er. And now, we see two-part verbs developing out of verb + prepositional phrases, FROBLEM 11.28: We ordinarily make agent nouns out of verbs by adding -er: rob > robber, fly > flyer, and so on. What are the agentive forms for these: someone who pick л up: Me is a . Someone who fixes up: He is a Someone who sets up: lie is a Comment on the claim ibul fix up is grammatically two words, that the past tense form; fixed upt RQlfix uppedy shows that they have to be considered grammatically шь synthesized.

The Impersonal A construction that has entirely disappeared in Mod£ is the impersonal sentence, a pattern that with certain verbs apparently lacked a subject: ИШ nones pinges ne iyste. (Him no things ne desired.) Me sceamab". (Me shames,} Him fryrstede. (Him thirsted.) It is clear from the case of the pronouns (bine—accusative, me- -dative or accusative, Aim—dative) that none of them can be the subject of the sentence. Two things happened to bring about the disappearance of this pattern. First, m Late OL, hit (it) began to occur more frequently in the subject position, particularly when the verb took а ьеШеш-с complement: Hit sceartiad me pat he me syh&, (It shames me that he sees me.) Hit licode him pat ic eode. (h pleased him that I left.) It is from that pattern thai we have ModE sentences such as ft seems that It appears that.,., it pleases me that The second cause of its disappearance

290

GRAMMAR AND SOUND

was in the loss of case distinctions between nominative and accusative/dative among common nouns. So long as the case of the common noun was apparent, it was always clear that the first noun phrase in such sentences was not the subject: psrrn gumttm hungrede. CThe men [dative! hungered.) But when these inflections were lost and the noun was in the singular (impersonal verbs always occurred in the singular), there was no overt indication that the first noun phrase was not the subject: ре татт hungered.

On the basis of the overwhelming frequency of the subject-verb order, the first noun phrase was natural!y reinterpreted as the subject. (The pattern endured into the sixteenth century in the limited structure meseems, metfar&S, me had rather,)

SUMMARY The major grammatical changes in Modem English are these: 1.

The Nominal system: a. Inflections are greatly reduced. b. Prepositions largely assume the function of the dative and genitive cases. c. At the develop as articles, l his j that - thesejthose as demonstratives, d. The rules for relative clauses are simplified. ~*ft

2.

The Verbal system: a. b. c. d. e. f.

3.

rt#>P^

jj

The auxiliary system greatly expands. M^ ™ f ^ r\ft Al! person endings except third person singular disappear. Many ablaut verbs become regular, Many regular verbs (as we shall see) become irregular, Do develops in negative and interrogative patterns, Many new verb + preposition and verb + particle patterns appear.

The Adjectival system: a. Comparison evolves from strictly inflected to mixed periphrasticinfieetion. b, Strong-weak distinctions disappear along with inflections.

GRAMMATICAL CIIANXJEI THE VERB PHRA3E

4.

The Sentence: a. Words order settles into S-V-(C), b. Negation precedes the negated. c. The impersonal pattern disappears.

291

Chapter 12 BETWEEN GRAMMAR AND PHONOLOGY

SOUNDS We have seen how semantics and grammar interpenetrate in the continually changing patterns of transitive and intransitive verbs, of new distributions for verbs iike grow, spread\ whisper, and tench. We have also seen how sound patterns and semantics merge in phonetic symbolism, creating words like teenyt boom, iinkle, and splash. And we have briefly seen how phonology and grammar have impinged on one another with some very important consequences to the history of the Language: When in Late OE, inflectional ending* on nouns began to level because they were so lightly stressed, the elaborate case and gsneter distinctions of OE began to disappear too. And as casedistinctions disappeared, English speakers came to reiy more and more on an already increasingly regular word order to understand how sentence elements related to one another. And this loss of inflectional complexity and greater reliance on word order very likely encouraged other periphrastic expressions to develop throughout the language, further weakening the need for inflectional distinctions. Phonology and grammar have interpenetrated in other, less obvious ways. When in the fifteenth century, syllabic inflectional endings on nouns and verbs were well on their way lo dropping their vowels in most contexts: 293

29f

GRAMMAR AMD SOUNU

caries > fans, wmysshed > vanished, the remaining consonants varied according to the kinds of sounds they occurred after. They assimilated, became like those sounds. With a few exceptions, voiceless word-final sounds determined voiceless inflectional endings: bank, banks, banked. Voiced wordfinal sounds determined voiced endings: bag, bags, bagged. Thus a minor phonological change resulted in a more complicated way to express grammatical distinctions. Other sound changes have resulted in more obvious grammaticLil and semantic differences. At one time, English verbs were, with some exceptions, dearly split between those with the regular dental preterit ending: name, named, named, and those irregular verbs with ablaut, or vowel changes: stag, sons, wng- B l l t in t a t e ME a number of once regular verbs joined the irregular group. Because of certain conditions which we shall explore in Chapter Fourteen, a sound change modified the vowel quantity in the past tense of many verbs from long to short. Then a second sound change altered the quality of Lhe vowel in the present tense, raising it to a higher vowel (review page 144), This created two different vowels in present and past tense verb forms. Thus from once regular verbs we now have the following irregular verbs, verbs whose irregularities are quite different in origin from the Indo-European ablaut verbs: feel-felt, slide-siid, bleed-Ыеб, speed-sped, meet-met, feed-fed\ menn-meanJ, hose-left, hide-hid. Had the first sound change- not occurred, we would pronounce the vowel in the past tense of these verbs as we do the ModE present: feeied. glided, bleeded, meeied, hided, and so on. (25, 28, 174) A yet more extreme kind of sound change has created what we now perceive to be completely different, though grammatically related words. In West Germanic, before the earliest written records, even before the AngloSaxons left the Continent, there was a derivational suffix that changed nouns and adjectives into verbs: *-jm. It gave the new verb a sense of bringing about what the noun or adjective indicated. Thus to the Germanic stem *blod(bloodj, the suffix *-ja» produced *bludjan, meaning to bring about or cause to produce blood. The *-j- in the ending represents a sound not (oo different from the first sound in yet. It caused the rounded back vowel in the first syllable, the loj in *btdd-, to umlaut, to be pronounced more toward the front of the mouth in anticipation of the front /y/ sound. The /6/ assimilated to the /y/ sound, giving a rounded front vowel we can write as /3s/. Not too long after this, the *-j- disappeared. Then in OL, the rounded front vowel unrounded to /c/. The whole sequence went like this: stem sU'in + affix fronting Jos&of *-junrounding *ЬШ- > *b!ddjan > *b!&djatt > bl&daii > Nidan We thus have ModE blood from btod and bleed from bledan, both from the root *blod.

BETWEEN GRAMMAR AND niOKODOOY

295

PROBLEM 12.1: The same kind of change gives us ModE pairs like food-feed, knot-knit, doom-deem, drop-drip, stunt-stint, foul-(de)fi}e, brood-breed, tooth- teethe. full-fill. Sound changes have created other pairs. Match up the pairs, which have been separated into groups (1) and (2): 1, 2,

slack, pass, beacon, holy, day, brass, grass, glass, prize, batch, breach, match, stick, stink, watch, web, Seam, do, blithe, fleet, rise, cafii. deed, make, wake, pace, praise, brazen, float, dawn, slake, bliss, graze, fore, weave, beckon, bake, rear, hallow, cool, stitch, glaze, stench, break. (97)

What kinds of meaning changes have also occurred? The pronunciation of words can also change from semi-grammatical causes when grammatical word boundaries are misunderstood. Thus in English, we have new!, nickname, and the now archaic mmcle because speakers mistakenly attached the -n from a»: an ewt, an ekename, an unck, to the beginning of the following word. The opposite occurred with earlier napron, nadder, nauger, and ггшщаге. The ModE orange, borrowed from French, differs from its earlier source, Spanish mrattja, because Old French speakets made the same mistake in the noun phrase un itaranja, PROBLEM 12.2: The same mistake has changed the form of a number of foreign borrowings. Two words, often an article and a noun, merged. Here are some borrowings. What are the ModE words? Spanish: el tagarto (the lizard), la reata (the rope). French: la crosse (the bishop's crook), dis mal (days unlucky), alas (oh, weary!). Dutch: de kooi (the cage—for catching birds), de affodil (from Latin asphodelus). German: garausl (all out!—i.e., my beer stein is empty). Arabic: al ikslr (the powder), aljabr (a shortening of a longer phrase meaning the reduction—to another form), al klmiya" (the extracting and mixing of juices}, a! kuhl (the powder), al qilt (the ashes of the salt wort), al matiakh (the weather), amir a! barh (commander of the seas—the barh was dropped when the word was borrowed by the French), el Hah (the god). In талу of these, the second word was borrowed by the Arabic from Greet. All except the last were borrowed into English through French, Latin, or Italian,

STRESS Unfortunately, a history of the most crucial area where syntax and phonology interact cannot be written with any certainty. When we utter a sentence in

296

GRAMMAR AND SOUND

ModE, the pitch of our voices rises and falls in intonation patterns. When we make a statement for example, our voices rise and then almost immediately fall to a pitch below that from which the rise began: э mo_ He went lo the | ~ ~ \ \ vies.

1 = low pitch middle pilch high pitch

PROBLEM 12.3: Plot the intonation contour for these sentences: 1.

He went home.

2. Me went there. Э. He went to it. 4, 5.

Did he leave ? Did he leave it ? 6. Did he leave it there ? 7. What did he leave? 8. What did he leave there? 9. What did he lease therefor ? 10. Did he leave or stay ? Use ordinary intonation. Can you predict grammatically when the pitch begins to rise and fall? As our voices rise and fall in these contours, we stress some parts of words more than other parts, some words in a sentence more than other words. Ordinarily, the pitch begins to rise on the word we stress the most in a phrase, and the word we stress most tends to be the last content word in a grammatical phrase or clause. He saw the [

MA~[

SAW He, t

1 know who he \

\. \to

them.

yesterday.

PROBLEM 12.4: What words bear phrase stress in the problem above? In utterances longer than just a few words, we tend to retard the rhythm of our speech at major syntactic junctures, usually rising one pitch level, from 2 to 3, and then Rilling not to I but buck to the original 2, still stressing the last content word of a phrase:

BETWEEN GRAMMAR AND PHONOLOGY

-

! NO Yesterday after Г ~ \ - o n .

' i

У*

297

J QU1 when everything was I \ et 2

й

2

з Оseveral of us from the Г ^Ч nice

'2

j4-

1

i LUw • 2

ent out to

2

12.5: Plot the pitch and stress patterns in these sentences. Numbers (6) and {7} have pitch patterns we have not described above. 1. 2. 3. 4. 5. 6. 7.

Unfortunately, everyone who was there on time decided to stay for lunch. My friend George, the guy that lives next door, wanted to meet you. If you decide you want it now, tell me about it. He was therefore unable to decide what to do. Therefore, we were unabk to decide what to do with all the money. George, Bill, Mary, and Nancy were all there. ! gave him a hook, a pent a piece of paper, and a piece of cardboard.

In all these cases, stress, pitch, and juncture clearly depend on part of speech and syntactic structure. Unfortunately, we can say relatively little about whether such rules may have been significantly different in OE or ME the te*mal evidence is insufficient (59, 70, Й6, 219) 12.6: We do know that many words we borrowed from the French were stressed on the last syllable: par DON, paLAIS (palace), hoNOUR avenTURE (adventure), graMAIRE (grammar). Here are some verses from Chaucer. Cummcnt on tbc stress patterns of the words in boldface, citing your evidence. J.

And songen, everych in hys wysc The mosle golenipne servtsc By noote , , , (And sung, each in his way The moit solemn service By note,..) From The Book ofihe Duchess, 11. 301-3.

2.

And Cleopatrc, with al ihy passyuun Hyde yt your trouthe of love and your renoun; From TntLefiend of Good Women, II- 259-60.

3.

This Somonour bar to hym a stif burduun, Was ncverc trompe of half so greet a soun. (This Summoner bore lo him a stiff burdoun (burden, or bass accompaniment to singing) Was never trumpet of ha if so great a sound.) From The Canterbury Titles, The General Prolujji1*, IL 673-74.

298

4.

GRAMMAR AND SOUND

Bui trewely to Iclkn atle laste, Hft was in chirchi: a noble ecclesiaste. From The Canterbury Tales, The General Prologue, 1!. 707-8.

We know from metrical evidence that compound words in OE and ME were also stressed on the first syllable. And from metrical evidence, we can be certain that the second root word of a two-word com pound was less strongly stressed than the main syllable, but based on spelling evidence, also more strongly stressed than any of its non-root syllables. Thus there were /

u

\

u

certainly three distinct levels of word-stress in compounds like tungerefa (district officer), beaditrof (bold in battle), and manslaga (murderer). Beyond this, we can say relatively little, particularly about the complex patterns of intonation, or about sentence and phrase stress. In MndE, three distinctions in stress seem to be fairly clear There is primary sines!;—the loudest stress we hear in monosyllables uttered alone or i

t

t

'

t

t

in the first part of a compound word: dog, топ, house, tree, flophouse, steambout, highway. The second degree of stress is most clearly heard in the second half of compound words: doghouse, steamboat, highway, hambone, footsore. It is also heard in some non-compound words: Plato, transcribe, remit, elbow, antique, Ff the vowel quality is clear, the stress is usually one of the first two degrees. A light degree of stress is more common in polysyllabic words, the one associated with the obscured {sj vowel: allow, except, awful, occur. All three stresses can be hard in housekeeper, accidental, takeover, sleighriding, transcribing, inducement, remitted, ecumenical. PROBLEM 12.7: Transcribe these words including stress: antagonisms,pacification, extrasensory, disestablished, car carriers, protuberances, poly dialectal, revisionism; entymohgy, raikaading, encyclopedias,floorcoierings,transportation, ielephonically. PROBLEM 12.8: What can you tell about sentence stress from the fact that ne kabban (ne have) was often written nabben; ne com (ne am) as neom; ne wylie (ne will) as nyUe; ne war (ne knew) as not? What can you tell about stress from the fact that on in gerundive phrases like on hunting was often written as a-huntingt and then dropped enMrely? What can you tell about stress from rhe fact that the Oli ich (I) became / pronounced like the short i in is, that art thou was often written arrow, hasj thou as hustow, wilt thou аз wiltou, that he and kit (he and it) were occasionally written just a, that kit became it; from the fact that in ML the perfect auxiliary hoot was written occasionally as a? What do you conclude from these sentences from Early

BETWEEN CBAHJUAR AND PHO14OLOGY

299

ModE: Tlfkast found me out, У'иге a scondrel, I/'is gone, Let's see't, He looked from one t'other, lie!e discover*t, Нее"Id telPm so'! Phonology has also influenced grammar indirectly when two patterns arc so phonologically similar that they could easily be confused and therefore cither began to differ or to fall together. The former may have happened when English adopted the Danish third person plural pronouns: they, them, and their to replace the OE hijbte, kin, and hira. Had the initial tk- sound not replaced the OE initial ft-, the third person masculine singular, the third person feminine singular, and (he third person plural would have become too similar to one another to be easily distinguished: iimiiiiiutiu'

genitive dative

Third person masculine singular:

he

hine

bis

Third person feminine singular:

heo

hi{e)

hi(e)re hire

hi(ej

hi(e)

hira j

Third person plural:

him him

The same underlying need to maximally distinguish easily confused forms may have contributed Ю the development of she to replace lien. Had it not, the M o d t masculine and feminine nominative pronouns would have been too similar to distinguish easily. Of course, this in no way tells us where schee came from originally. It seems to have originated in the Northeast Midlands щАл twelfth century. (209) The opposite occurred when the similarity between the OE and ME progressive ending -endej-ande and the gerund -uttge ending, caused the two to merge, ultimately leading, perhaps, to я much wider use of the progressive. In this case, the structural load carried by the two was not great enough or distinct enough to prevent Lhem from falling together. Those sound changes which most profoundly affect the phonological structure of a language, though, usually occur independently of the grammar of that language. When a sound changes—or more accurately—when on certain occasions certain speakers begin to use one sound in a word rather than another sound—that sound will ordinarily change in all words, regardless of whether they are nouns, verbs, adjectives, or prepositions. And unless there are predictable conditions or unpredictable analogies, the sound will change in all grammatical contexts. Therefore, the discussion in the next chapter on the ways English sound patterns have changed will largely ignore grammatical and semantic questions.

Chapter 13 PHONOLOGICAL CHANGE

TFTE IMPORTANCE OV PRONUNCIATION

Linguists and fifth graders alike have labored over a problem that originated sometime during the Middle llngliah period. It was then that sound changes in the language began to lay the groundwork for our present orthographic confusions, so that today the way we spell words no longer accurately represents how we pronounce them. In most other European languages, spelling is more regular. Those who grow up speaking Spanish, German, or Italian, can Jearn to read and write those languages with fewer of the problems facing an English-speaking child. Those linguists who try lo discover how Hnglish was pronounced as little as two centuries ago wrestle with many of (he same problems: The "fit" between English orthography and pronunciation is at best imperfect and occasionally chaotic. The ways in which our speech has changed ean shed some light on these problems. But the way we pronounce words is important to more than just understanding how we spell. Perhaps more than any other characteristic, our accent assigns us to a geographical area, to a social class in that area. It tells a listener whether we are being formal or informal, casual or intimate. We judge and are judged in turn by how we pronounce our words. Indeed, pronunciation 301

302

GRAMMAR AND SOUND

has been a social criterion at least since the first quarter of the twelfth century, no doubt earlier. ГГ we can understand something about how these judgments have come about, about, the history of our attitudes toward our pronunciation, more importantly about how our pronunciation has changed, we shall perhaps better understand our own responses LO those who speak differently from us. Because spelling is a major source of evidence of sound changes, we haveto understand not only how it has changed in the last 1500 years, but also the principles and history of writing and how they relate to phonology. Fit is how well the letters in an alphabet match the significant sounds of a language. An alphabet can have too many symbols1 for the number of sounds: , , , , and " h a r d " , for example, can all represent a /k/ sound; clique, kid, chorus, sick, cap. Or an alphabet can have too few symbols to represent sounds: English, for example, has no unambiguous single symbol for the /z/ sound at the end of rouge or the middle of leisure; or for the sounds represented by (th> in ihhi or the <sh) in ship. And perhaps worst of all, symbols can represent sounds inconsistently: Compare in cough, though, tough, hiccough, through, and ought.

THE PHONEMIC PRINCIPLE

PROBLEM 13.]: Here arc some words which illustrate the significant sounds in a common dialect of American English, the one spoken around the Great Lakes, and extending more west than east: dot, pub, crag, this,quit, keg, sad, fawn, wool, thing, hire, chose, shame, rouge, joy, year, wow, zip, ax. (1) What letters could we drop from the alphabet? (2) What sounds have no individual letter to represent them? (3) What are some of the obvious inconsistencies between letters and sounds? PROBLEM 13.2: Listen carefully to the sound in these words: eighth, ten, Bertram, button, butter. Say eighth, but when you get to the /t/ sound, hold your tongue in that position and say train. По the same and say team. Arc the three sounds the " s a m e " ? Are the ,'t/ sounds in button and butter the same? How do they differ from the /t/ sounds in (rain and team? How maity perceptibly different sounds are represented by which we still call /t/ 1 We shall us* Ihrce natations in this chapter, Angles, < >, will endow leriers of the alphabet. Slashes, ,' /.will enclose cafcaorics of sounds. Square brackets, [ ], will enclose specific sounds wLihin a category, a concept which will be fully explained in the next few paragraphs.

PHONOLOGICAL CHANGE

303

sounds? IT we require a different symboT for every different sound, why do we not require s different symbol for these different /t/ sounds? If we listen closely to the sounds we make, we quickly realize that the number of different sounds we can hear exceeds the Idlers in the alphabet. Indeed, with a little training, we can distinguish literally hundreds of different sounds. But in ordinary listening, we consciously discriminate considerably fewer. Even though physiologically, we produce them in different ways, the /t/ sounds in Bertram, eighth, and tea are psychologically the "same." Yet even though we unconsciously articulate these /t/ sounds more differently than we do the initial sounds in sip and ship, we consciously discriminate them mueb less readily than the initial /s/ and /S/ in sip and ship. Actually, all sounds differ. For at a sub-molecular level, no two events or objeets are precisely alike. As the differences become more gross, we can, if we sharpen our cars, distinguish phonetic characteristics that an untrained observer would not detect. But for the purposes to which the sounds are put, they might still be considered functionally equivalent. It is at this level that the /i/ sounds are functional "sames," or more accurately, are responded to as functionally equivalent members of the same class of /t/ sounds. We respond to the critical class characteristics rather than to the individual differences within the class. 13.3: At a somewhat different level of analysis, all /t/ sounds are not entirely functionally equivalent. How do we recognize a foreign accent*? Does such a speaker substitute different sounds for native sounds4.' If he does, how do we understand him? PROBLEM

At some level, however, differences do become important enough to make us distinguish sounds as belonging to different significant classes. The ordinary speaker has thresholds of sensitivity for different sounds that require him to distinguish differences according to the rules of the language, rules that he has subconsciously learned. An English speaker must learn to group the very different /t/ sounds into one class, to discriminate them from a similar class of /d/ sounds. He must learn to distinguish the class of/s/ sounds (as in sip, swing, and bus) from the slightly different class of jij sounds (as in strip, bush, and mission). The minute differences among the members of these sound classes, however, are not random. We can predict that for a /t/ in Bertram, we will retract the tongue slightly to adjust to the fij sounds that surround the /t/ sound. In eighth, we push the tongue forward to make a /t/ sound, because the following segment which we spell and pronounce /A,1' is made with the tongue between or against the teeth. Each class of significant sounds, all the sounds which taken together as a group we perceive as "sames," are called phonemes. The individual members of a phoneme, of the class of

ЗЬ4

CrRAMMAR AND

sounds, are called allophonre. Phonemes are written between slant lines, allophones within square brackets. Some of the allophoues of the phoneme /d/r for example, are these; [d] A retracted sound that occurs before or after /r/, as [n droop. [dl A fronted sound that occurs before or after sounds spelled
59,73, 8+, «6,219) PROBLEM 13.5: Here is a sample problem in an imaginary language. How many phonemes are in this language? The letters represent sounds as in English, /g/ as in gun, ftj as in pep, jot as the [o] in pope. [pemp]-boy [ijebepj-inan [kont]-house [kogomp]-tree

[font]-liill [fenze]-sun [sezokj-dog [poge]-hand

[tovemf]-cat [pombo}-duck [sezes]-river [kondej-apple

[kevek]-door [tontl-cloud [pedek]-foot [kemve]-nose

This language has two vowel phonemes: /o/ and /e/. We cannot predict where cither will occur. The language has six consonant phonemes. (1) There are two allophones belonging to a phoneme we shall symbolize as /N/: [n] and [m]. [n] occurs only before sounds made where [n] is made—on the

-

PHONOLOGICAL CHANOt

305

alveolar ridge behind the teeth, before [t], [d], [я], and [z]. The [m] allophone occurs only before sounds where it is made, at the lips: [p], fb], [П, and [v]. They are roughly alike. Neither occurs where the other does. When one of them does occur, its environment is entirely predictable. Therefore we have one phoneme, which we can symbolize as /N/. The other sis segments can be predicted according to where they occur in a word- The voiceless sounds, sounds made when the vocal cords are not vibrating, [p], [t], [k], [f], and [s], occur only at the beginning and end of words. The corresponding sounds, [Ъ]. [d], [g\, [v], [/], occur only in the middle of a word. Roughly alike are [b] and [p), [d] and [t], [g] and [к], [Г] and [v], [s] and [z]. Neither occurs where the other does. They constitute the pairs of allophones in phoneme classes which we can arbitrarily symbolize as /P/, /T/, /K/, /F/, and /S/. We can now write the first five words in the problem like this! /PeNP/, PePeP/, /KoNT/, /KoKoNP/, /FoNT/. PROBLEM 13.6: Write five more of the words in phonemic notation. Translate these into allophonic notation: /PcKeNP/, /KoKoK/, /TcPoNT/, PROBLEM 13,7: Suppose by a process similar to what happened in ME, final vowels were lost in this language. PROBLEM 13-8: Suppose that \p] after [m], and [tl after [n] were also lost. Suppose just one were lost. PROBLLM 13.9: Suppose that before these final vowels and consonants were lost, these words were borrowed into this language: [vedcp], [zogop]> [vovof], [zoiokj. The rationale behind organizing individual sound segments like these into classes of sound segments is three-fold. First., phonemes may have a psychological existence. Asked if keep and c<W begin with the " s a m e " sound, English speakers will answer yes. Asked if keep and cheap, or cool and (ool begin with the same sound, they will answer no. Speakers of certain other languages will answer differently if Lheir phonology organizes their phonetic segments into different classes. Second, the overall patterning of sounds in a language can be better perceived if we study how phonemes rather than individual phonetic segments are organized in a language. As we shall see, English phonemes, the classes of sounds, arrange themselves into a highly symmetrical relationship that would be obscured it' we simply cited all the allophones of English, 'third, it is more economical to represent sounds phonemically than with individual phonetic segments. English requires fewer than 35 symbols to represent its pronunciation, if we choose to represent that

306

GRAMMAR AMD SOUND

pronunciation phonemically. If we choose to represent it with phonetic segments, we would need as many symbols as there are sounds our ears can delect, and that number reaches into the hundreds. (81) Modern linguistic science has probably expended more effort in the last two centuries on phonology than it has On any other aspect of linguistic research. The nineteenth century was distinguished by those Scandinavian and German linguists who successfully systematized the regular phonological relationships among a wide variety of related languages across Europe and parts of Asia, the group ofLangjagcs that has come to be known as the IndoEuropean family, fn Hi is century, linguists have developed elaborate techniques to describe the phonological structure of modern languages. In the last decade, generative linguists have extended the bases of generative grammars to the description of deep and surface phonological structures, (33, l(Ki) Although generative phonologists have described F.nglish sound patterns with considerable generality and have contributed greatly to the study of historical English phonology, it woutd require too much space to develop the needed background in this test. Therefore, we wilt rely here on a more conservative theory of phonology. (219)

THE PHONOLOGY OF MODERN ENGLISH

Tbe place to start in a traditional description of English phonology is the physiology of the vocal apparatus. (See Figure 13.1.) Although each part of the anatomy that helps us talk functions primarily in some other capacity (the teeth to bite, the tongue to swallow, the diaphragm to breathe, and so on), we have evolved until those separately functioning organs now also integrate into a complex speech mechanism. As we contract the intercostal muscles along the ribs in short pulsations to push the air through the pharynx and past the glottis, or vocal cords, we allow the cords to vibrate, producing a humming sound» or not. This gives us two large categories of phonological segments: voiced and voiceless sounds. The vibrations of the vocal cords can be felt by placing the thumb and forefinger on either side of the adam's apple and alternately pronouncing ss-zz,ff-vo, shk-zhfik 13,10: What are the voiced segments in Problem 13.1? What are the voice Less segments'/ Voicing is an important phonological environment, Tn many languages, including the artificiai one in Problem 13.5, segments at the beginning and end of a word arc voiceless, while segments between vowels arc voiced.

PHONOLOGICAL CHAI^Gfc

307

Figure 13.1, The Physiology of the Vocal Apparatus Consonants

Once the column of vibrating air readies the oral cavity, its harmonics arc modified by the tongue, teeth, tension of Lhe muscles in the mouth я rid throat, and the volume of the cavity. If there is little от по constriction in the oral cavity, then vocalic or vowel sounds are produced. If there is constriction, then non-vocalic sounds. Where and how much we constrict segments allows us to define nonvocalic sounds more exactly. The consonantal sounds can he defined by whether we completely stop the air, then release it, giving a group of sounds called voiced and voiceless stops: ;p/, /b/, /t/, /d/, /k/, ,'g/ (as in gun)\ от whether the air is aliowed to cseape while the sound is produced, giving us a class of c&niinuaiits. The continuants can be subdivided by a number of features. If we constrict the passage enough to produce audible friction, then we have a class of voiced and voiceless spirants: ;'f/, /v/, /B/ (as in thin), {6} (as in this), /s/, /z/, /s/ (as in she), ffl (as in leisure). A combination of a brief top and a spirant is called an affricate: /j/ (as in judge) and /ё/ (as in church)*

30S

GKAMVAR AMD BOUND

Tf we let the air escape through ihe nasal passage instead of through the mouth, then the continuant sounds arc nasal segments: /m/t /n/, and the sound we usually spell as in sing or as before or , as in think and finger, a sound symbolized /g/. In the eon text of nasal consonants, vowels take on nasal coloring. In some English and American dialects, the vowels become so nasalized thaL the nasal consonant disappears: [шж] for man, for example. Nasals also tend to obscure the quality of a vowel, often changing it. In some dialects, for example, /e/ before /n/ tends to become /i/, making pen and pin homonyms. Because these next sounds are so full of acoustic resonances, they have been classified by some as having both vocalic and consonantal characteristics: /)/, a lateral made with the air escaping over the side of the tongue, and /r/, a retroflex sound made in a variety of ways, one by humping the back of the tongue up close to the roof of the mouth for medial or final /r/ sounds, another by slightly curling back the tip of the tongue for initial /r/ sounds. Three segments, semi-vowels, arc very similar to vowels in their quality of resonance: The /y/ in yes and boy is close to the vowel we spell <ee> in eel. The same is true with the initial /w/ in wow, which is close to the vowel we spell as in ooze. The third semi-vowel is ,/h/. Always voiceless, it takes on the articulatory shape of the following vowel: heal, head, hat;, hut, hot, hoot. You can demonstrate this by starting to pronounce the /h;' in heat but saying hot instead, /r/, /I/, /y/, and /w/ strongly influence the vowels around them because they are so much like vowels. For example, ME historically symbolized the " a h " sound, /a/, as in father. By the sixteenth century, this sound changed to an /a;/ sound as in cat. But in the environments of/w/, /r/, and /1/, various changes occurred in various dialects: Before /r/, it moved back to /a/, as in barn or start; after /w/, it also moved back to /a/: quad, wad; between /w/ and /r/ or before /I/, it changed to a vowel even further back: war, quarrel, call, matt, Шк, The point in the mouth where the speaker most constricts the articulation is the last criterion. Constriction at the lips gives bilabials; /p/-/b/-/w/-/m/. Constriction between upper teeth and lower lip gives laMo-dentals: /f/-/v/. Constriction with the tip of the tongue between or against the teeth gives interdental;: /6/ as in thick and /Й/ as in ihis. Constriction with the tongue on the alveolar ridge gives alveolar»: ftj, /d/, ,'s,', /z/, /n/, and /1/. Constriction by placing various parts of the tip and blade of the tongue behind the alveolar ridge just at the edge of the palate gives aheo-palatals; /]/, /£/, /z/, /s/, /y/ and ft}. Constriction at the back of the tongue with the soft velum gives the velars: /k/, /g/, /n/, /h/ is gluttaL The consonants are arranged schematically in Table 13.1, PROBLEM

13.11: Transcribe the following words phonemically. Use the vowel

PHONOLOGICAL CHANGE

TABLE 1ЭЛ

A SCHEMATIC REPRESENT AT] ON OF THE CONSONANTS

Bilabial Stop vcls vcd Spirant vcls vcd Affricate vcls vcd Nasal Lateral Rctroflex Semi-vowel

309

Labio-dcntfil

Interdental Alveolar

ftt

/P/ /b/

M

M

/e/

/8/

N

Ы

M /1/

Velar

N

Ш

/d/

/f/

H

Alveo-paL

m /J/ W /y/

Ш Glottal /h/

symbol /i/ for the letter , the vowel symbol /e/ for the letter <e), and the vowel symbol /as/ for the letter . tip, sin, fill, wet, vend, spend, stack, chest, gyp, thick, lapsed, chin, chink, gem, gym, swing, yelped, hung, ist Гhat, thanks, hanged\ flange, stripped, script, quit, squelched, glimpsed, hinge, limb, damn, baths, shrink, hatched.

Vowels The vowels are classified differently from the consonants. The first distinction is between simple and complex vowels, or between monophthongs and diphthongs. The monophthongs, or simple vowels, have only one major segment, as in the vowels of pit,pet,pat,putt, pot, put, and ought. Diphthongs have two clearly perceptible segments in boy, house, and by; less clearly perceptible in eat, aid. food, and mad. We produce vowels by shaping the oral cavity with the tongue, so the categories of simple vowels can be defined by where in the mouth we bunch the tongue. Three vuwcls are front Towde: ft) as in bit, /«/ as in bet, and /a;/, called asch, as in bat. Two vowels are central: /s/, called schwa, as in but or allow, and /a/ as \n father. And three are back vowel»: /u/ as in put, foj as in the Eastern New England pronunciation of boat, a sound not often heard any more, and {of, called open-», as in bought. These vowels can also be classified by height: /i/ and /u/ are high; /e/, fa/, and /o/ are mid; /я/, /a/, and /з/ are low.

310

GRAMMAR AND SOUND

Л characteristic of the back vowels is lip rounding. Unlike a language like German, rounding is not a significant feature of the English vowel system, whose main contrasts arc between back-front and high-low. No two vowels contrast only because one is rounded and ihe olhcr not. German, on the other hand, has two rounded front vowels: /u/ and /o/. They roughly correspond to /i/ and /e/ in frontness and height, but are produced with rounded lips. They contrast with German /u/ and /o/ which are also round, but back. Using a threc-by-three system, we can classify the simple vowels like this: Front High

Mid Low

Central

l

Back u о

e a

0

Other systems attempt to represent the shape of the oral cavity more graphically and use something that more closely resembles a vowel triangle; i

u e

э at

о э

a That is, as physiologically /эе/ sounds are both more front and higher than /a/ sounds, so the vowel triangle represents that fact. PROBLEM 13,12: Transcribe these words using the system outlined above: pods, coughs, runs, puts, canned, checked, squashed, gnats, axed, odds, as, clucked, width, chinks, jawed, thwacked, strong, this, whizzed, cents, sieves, wronged, chalked, throngs, munch, scratched, crunched. PROBLEM 13.13: Consonants can be conditioned by voweb as well as by other consonants. How do the allophoncs of /к/ differ in keep and cool'} Why do they differ? Consonants can also condition vowels: How do the altophones of/a/, /i/, /se/, isj, /u/, and /e/ differ in these words: hot-hod, bit-bill, but-bang, bought-bawd, put-pull, pep-pen. Why do they differ? Diphthongs have two distinct parts, a nucleus and a glide. The mosL obvious diphthongs in English are in buy, bough, and boy. They begin with a stressed nucleus then glide off into a more constricted vowel; /bay/ от /bai/, depending on how we choose to rcprcsenL the sound: /baw/ or /bau/; and /boy/ or /bsi/. How we choose between /an/ and /aw/, /ai/ and /ay/, /si/ and

PHONOLOGICAL CHANGE

311

/эу/ depends on considerations which need not concern us here. For our purposes, we shall use /i/ and /u/ to represent the ModE offglides. The less clearly perceptible diphthongs in Lnglish are the vowels in these n r words: (1) beer, /bit/; (2) bait: /bet/ /belt/; (3) frw/: /but/; (4) btfaf: /hot/ or /bout/. The way we use symbols to describe these sounds can become complex because some systems of descriptive phonology use symbols for long vowels to represent what in ModE arc diphthongs. But strictly speaking, length alone is not a significant feature that distinguishes one category of English sounds from another. Everything else being equal» diphthongs are temporally longer than monophthongs. But we do not use length as (he feature which distinguishes diphthongal beef from monophthongs] bit. It is rather an off-glide and a quality of tejiseiiess in a word like beet not found in the more lax and monophthongal pronunciation of bit. [You can detect this tension by holding the muscle below your jaw directly under your tongue and alternately pronouncing beat-bit, pool-put, bate-bet. The muscles tense for the first sound, but not for the second.) In this study, however, we will use the unitary symbols to represent these latter diphthongs: /I/ for the vowel in eel, /e/ for the vowel in ail, /0/ for the vowel in pool find /o/ for the vowel in pole, because in OE, ME, and Early Modern English, these were still probably long monophthongs.

ё/е

u/u o/o

PROBLEM 13.14: Transcribe these words: chides, choice, squeezed, skighf, plowed, speeds, horse, cleared, pears, joked, voiced, through, pierced, though, psyched, veins. Are there such things as triphthongs? Diphthongs arc by far the least stable category of sounds, Tn the history of English, they have appeared find disappeared more frequently than any other kind of segment. A great variety of diphLhongs other than the ones we have been examining here can be heard from Eastern New Englanders, New Yorkers, Cincinnatians, and so on, a fact which testifies to the geographical biases of this text. PROBLEM 13.15: Find someone raised outside the dialect area we have been illustrating and describe some of his diphthongs. One characteristic of English phonology requires special, if too brief mention here. The schwa symbol, /э/, has been used not only for the vowel sound in hut^ /b&t/, but also has been used with /r/, /1/, /m/ and /n/ to symbolize the syllabic quality of the final syllables in words like bother, barret, bottom, and button; and the stressed vowel quality in words like burr, sir, and

212

GRAMMAR AND SOUND

her: /boSar/, /beral/, /batem/, /baton/, /Ьэг/, /гаг/, /h^r/. Л more exact phonetic transcription has a special symbol for Ihe /эг,1' sound: [эг], and for the stressed /э/ it lias /A/. Thus £wf can be transcribed /Ьл1/, while allow is transcribed /sdau/. We shall use ;'э/ in /эг/ and instead of /ff/. Under light stress, the vowel quality of most syllables is also symbolized as /э/, though sometimes iL may tend toward [ij or [i] (a central high vowel), as in thy last syllable of

fined. PROBLEM 13,16: Transcribe the following words: firebirds, bottomed,rotten, squirrels, earls, fathomed, woolens, squirters. PROBLEM 13.17: Transcribe the sentences in Problem 12.5 into phonemic notation, including stress patterns for individual words.

THE DEVELOPMENT OV WHITING The spelling system that Western Enropeans use to represent all these sounds has descended through Latin and Greek from a Semitic system dating from perhaps the seventeenth century в . с , a system which may have had its conceptual origins in the hieroglyphics of ancient Egypt, dating back to с 3500 в,с. Its successive adaptations from language ю language to fit different phonological patterns makes its history a complex one. The most primitive way to represent an idea, the way that certainly must have led to the invention of writing, is in pictographic representation. Virtually all cultures draw pictures for purposes ranging from telling a story to magic. Many have adapted their pictures to sending or leaving messages. The simplest picture represents in one frame an entire semantic complex. Traffic signs use this system: a picture of a deer or a child represent warnings we can instantly translate into ideas without using intervening words. More complex pictures represent more complex messages:

No single element in this message need specifically represent or be represented by any word, though a word can be attached to each clement. The- total picture simultaneously represents the message. Because there are several writings systems whose exact origins are unclear, it is impossible to say how many times, onee or many, it was realized that part of a picture could represent not only an idea, or a physical referent,

PHOKOLOOiCAL CHANGE

31.1

but the word associated with that referent as welT. But at least one person in the history of the human race recognized that a picture like represented an idea that need not but might be verbally translated into words, into something like Л deer is running. A picture like this can then be abstracted into components which,, by convention, might be attached to words associated with its idea or referent. We could, for example, abstract out of this picture a symbol for the word deer and a symbol for the word running and write it left to right in English word order;

Graphically, the change from pictographs to ideographs looks like this: Pre-writing Words

__ •r Ideas

Ideographic writing , * , Pictures ~* (symbols) -*• Words ->• Ideas

Pictures Unfortunately, while this intellectual leap equals any in man's history, a writing system based on ideographs is so unwieldy that only a few members of a culture ever havi; the time to learn it. Every distinct word needs a different symbol. The problems the Chinese are now having in achieving national literacy are so acute that they are in the process of changing their largely ideographic writing system to an alphabetic one. But they are able to step over an intermediate stage in the development of writing that our ancestors had to pass through. That next stage was the development of a syllabic writing system. Syllabic writing systeins arise when particular ideographs become associated with the sound of a particular word, and the symbol shifts from the idea the word represents to part of the word itself. The process is identical to the way we create a rebus:

WJV

• I see cartons of milk.

But this development can occur in two ways. The less efficient way, the way chosen by the Sumerians, was to have a distinctly diflcrent symbol for every different syllable. It would be as if we had a completely different symbol for each of these syllables: /ap/, /sap/, /tap/, /step/, /spat/, /pats/. Although there are many fewer syllables in a language than there are words, auch a syllabary would still be exceedingly large. Another way was devised by the Egyptians and then developed by the

314

(jRAMMAR AND SOUND

Semitic tribes. Because syllables in the Semitic languages are very simple—• often consisting ofjust CONSONANT + VOWEL, the Egyptians developed a syllabary (to complement their ideographic hieroglyphics) that represented only the first consonant of the syllable. It would be as if the ideograph

<ЯЕД

represented

/bi/, /be/, /ba/, /bu/, /bo/ and. soon. Tfaevowel was

not specifically identified. This was a great advance over theSumcrian syllabary because it isolated one symbol for one sound plus a very limited number of possible second sounds. Since the number of consonants in a language is much smaller than the number of syllables, a syllabary of this sort is much more efficient than one with a distinct symbol for every syllable. PROBLEM 13.18: Develop both kinds of syllabaries for ten ModE syllables out of an ideographic system of your own device. Then conventionalize the ideographs for easy writing. Why is a syllabary on the Semitic model so much more difficult for a language like English? Choose one of these writing media for your symbols: stone, wood, clay, paper. What difference would media mate? When around the tenth century B.C., the Greeks adopted this Semitic (or as they called ii, Phoenician) writing system, they improved on it in one important way. They recognized how useful it would be to have sounds for both consonants and vowels in their language, whose syllables were more complex than just the CONSONANT + VOWEL of the Semitic languages. They used the consonant symbols representing Semitic sounds that did not occur in Greek to represent Greek vowels. The first symbol of the Semitic alphabet, 'alcph, represented a syllable beginning with a glottal stop consonant (like a catch in the throat), a sound useless to the Greeks. But they adapted alcph to represent the vowel now known as alpha, our to represent ,'a/. The instant this change was accomplished, the Semitic syllabary became our modern alphabetic system, a system even more efficient than a syllabary. For now there is no ambiguity about how a syllable is to be pronounced and hence what it is supposed to mean. Thus beth, from which beta and our derive, no longer stood for any possible syllable beginning with /b/, but by default for only the single segment /b/. Any vowel following /b/ would be overtly represented by its own symbol. This system was then brought to Southeastern Europe and the Italian peninsula and passed along to various J tali с cultures. The Romans modified several letters to meet the needs of Latin phonology and then passed it on once mure to the civilizations they dominated and later to those Slavs now in Poland, Czechoslovakia, and Rumania who followed the Roman Catholic

PHONOUJQICAL CHANGE

SIS

Church. The Russians, Bulgarians, and Serbians adopted Lhc Greek-Cyrillic alphabet from Greek missionaries. When the Germanic invaders succeeded the Roman legions in fifth century Britain, they brought with them an alphabet they had probably borrowed from some northern I ta!ic culture in the second or Lhird century в , с , an alphabet similar enough to both the Roman and Greek alphabets to indicate its derivation, but distinctively different both in form and in lhc order of the letters. The Germanic tribes called it mnic writing, or secret writing. The letters, called runes, were believed to have magical properties. The alphabet was called the fuoarc after the first six letters: V Г] |> к f> V The letters were very angular, presumably because they most often had to be cut into stone or wood. A good many runic inscriptions and documents survive, the most famous perhaps being passages of a religious poem, "ТЪе Dream cf the Rood [cross]" engraved on the Ruthwell Cross, a large stone monument erected in Dumfriesshire, Scotland, around the end of the seventh century or beginning of ihe eighth. When the Christian missionaries from the Continent converted Britain in the seventh century, they naturally enough brought with them their Latin script, which quickly replaced the pagan-tainted fuSarc. But when somewhat later the Irish missionaries replaced Roman missionaries, they brought with them a modified Latin script called Insular script, which became the form of writing used in most OE documents. The later Anglo-Saxons, however, did adopt two symbols From the runic JTuoarc: After earlier using the digraph (two letters representing one sound) to represent Grtxk words with theta, [9]. they replaced it around A.D, 900 with the rune O ) , thorn, which represented either [9] or [6]. They also borrowed теулп, <р>, which at first represented the sound now symbolized by <w)r They also adapted a Latin letter, , with a line through it, <S>, and called it eth. It also came to represent the two interdental spirants, <S> and

щ Other major differences in the inventory of OE letters from ModE included the presence of the Latin ligature (two letters written as one) called asch: , first used to represent a high front sound close to the /y/ in yes, later some other sounds, all usually represented in modern O t texts with (g>Rarc or entirely absent from the OE alphabet were the letters for , , and <w>. was infrequent. Originally, the English used was also introduced later by Norman scribes to transliterate OE . <x) occurred in OF, but only in foreign words. The Norman scribes were also responsible for introducing (or reintroducing) various digraphs: for fbj and /в/j <sh> for /£;';

316

GRAMMAR AND SOUND

when pronounced ffli for in words like hwil-white. By the end of the Mid die English period, under the influence of continental scribal practice, the symbols , ,

, and

had disappeared and a new set of spelling conventions had replaced them, , , and (w) had aiso been introduced. Not until the 16th century, however, did and come to be distinguished from one another Earlier, for example, words like ever and until were spelled euer and vntli. <j> was originally an elongated version of
SOUND AND SPELLING In order to understand how the relationship between spelling and pronunciation has evoked and how to interpret older spellings, we have to distinguish between two ways that spelling can relate to sound: By symbols and by markers. The symbols directly represent a sound segment. Some symbols are simple: (t>, , and (p) in tap each represents a significant class of sounds: /tap/. Other symbols arc complex digraphs: (sh) represents /s/; represents either \Щ or /в/, for example. Among the vowels, <ec), , <ei>, (ie)> and so on, can all function as units to represent diphthongs. Markers do not directly represent their own sounds, but rather indicate the sounds that preceding or following symbols stand for. The "silent-e" in tide is a marker indicating the or <e> after a are both symbols and markers: They symbolizethe /i/ or /e/ sound in city and eel! and mark the (c> as is!. after would indicate both their own vowel quality and mark the as /k/; cut7 cot, cat. The same letters after generally mark whether it is to be pronounced /g/ or /j/: gasp, got and gum vs. gem and gin. Double consonants perform the same double function ."They redundantly mark their own quality and mark the monophthong before them: biter vs. bitter. The double consonant convention reflected phonetic reality in Early MF, because in OF, double consonants were actually pronounced longer than single consonants. And in M E, short vowels inevitably occurred before doubled consonants. The silent-e convention arose in Late ME, Because many words with a final <e> had long vowels preceding, there arose the practice of adding <e> to indicate any long vowel, even where the <e> did not etymological!y belong: ham > home, pin > pine. On the other hand, final <e> in some words of

PHONOLOGICAL CHANCE

317

Latin derivation do not mark a diphthong: doctrine, infinite, private, resolve, obdurate (though Milton dropped the in all these). Other letters which are neither symbols nor markers arc merely excrescent. The (h) in debt, the in hour, the in iamb, the

in psychology serve no phonetic purpose. Some Setters, though, only appear to serve no purpose: The (n) in condemn, hymn, and damn seems to be useless, but it represents a phonetic segment that is realized in derived words: hymnal, condemnation, damnation. The same is true with the
SJS

GRAMMAR AND SOUND

RECONSTRUCTING THE PAST We can now consider how we deduce from different binds of evidence the way English speech and writing have related to one another through the last 1500 or so years. To understand these changes, we have to understand how linguists reconstruct sound systems, what kinds of evidence and logic they use when they try to describe how an Anglo-Saxon would have pronounced a word fifteen centuries ago. Start with some assumptions: When a sound changes, it changes in similar environments in the same way unless special circumstances dictate Otherwise. That is, if an /a/ sound changes to an /я/ sound, then all /a/'s ordinarily change to /o/'s in all similar environments. Inconsistencies can be explained in a number of ways: Kirst, a word that still has, for example, an /a/ instead of the expected /л/ might have been borrowed from a dialect in which the sound change did not occur. For example, certain words spelled with , indicating that an /a/-like sound was changing to something more like an /o/-like sound. Thus OE hat became ME hoof, which became ModE whole. But the word ftaJet as in hale and liter; \\ also descends from lial. The :/л> in halt contradicts the generalization that /a/-like sounds became /o,'-like sounds. The explanation is that standard English, generally descended from East Midland dialects north of London, borrowed hate from the more Northern dialect areas where the did not change to ( o ) . PHOBLLM 13.20: Since we know hate was borrowed, can we conclude that it was borrowed before or after the change from to ? What information aboutdatingchangesdosuch faelsgive us? What date do we want concerning haki Second, a word that still has a segment that should have changed might be based by analogy on another form of the word in which the change did not occur. The /c/ in chosen, for example, is based on the analogy of the present tense pronunciation of choose. The original perfect participle in OR, gecoren, had a hard /k/ sound, quite in accord with the phonological rules, and would not have changed to .'е7. We "should" now pronounu; the perfect participle of choose as something like /кигэп/ or /karsn,1': fit has cooren it. Third, a sound ordinarily lost or changed might be reinlroduced into a word because of a spelling pronunciation. Boik was for many years rhymed with watk^ both having lost their ffl before /k/ in accord with a general loss of /I/ between /s/ and /k/. But it is not uncommon to hear к now pronounced /bslk/. Or a sound might be inconsistently changed by false etymology: The in author and throne was introduced by ME scribes because they thought

P H O N O I J O G I C A L CHANGE

319

they were Greek words and should have the spelling. The Old French words were autmir and irane. kinaily, some words just defy екр!а nation because we do not have enough solid evidence to decide on any answer. Most words spelled <ea> an; now pronounced /i/: неок, read, meat, seat, etc. But four arc pronounced /£/: steak, break, great, and yea., and several are pronounced jef: head, bread, death, breath, e t c All were at one time pronounced /E/. We can invent explanations; The four /e/ words were borrowed from a dialect in which /с/ did not change to /1/. The ft} words resulted from a sound change that often shortened /e/ to /e/ before alveolar or dental sounds. But finally, these are ad hoc statements made simply tft account for exceptions. But unless we assume that sound changes are basicaily regular, we cannot hope to generalize about phonological change. We would otherwise be left with only a mass of random changes. Both historical evidence and our aipirations to formulate scientific "laws" deny this as a possibility we can seriously consider. 13.21: In some cases, it is possible to reconstruct hypothetical forms of earlier pronunciations even when evidence from writing is not available. Imagine a cataclysm has cut off communications around Ihc world for 500 years, destroying all forms of writing that might have recorded pronunciations both before and after. Here is some evidence about pronunciation collected from languages for the year 2550. We arc fairly certain these languages are related and that each set of words all descend from some common ancestor word we are trying to reconstruct. Argue for your interpretation of the data.

British I. /drj 2, dap 3. zdik 4 guk 5. gen 6. bust

Irish

Amortcan

dri dap

tri tab

zdik

KLlg

guk gen

kug ken pozd

bost

Can aili.lM

tsrl tab satig kug ken posad

Austra- New Jamai- BermuMan Zealand cr can dan ti ri trt tri ap ta tap trjp sli stek stek stek kok kok kuk ku ken ksrj ken ken post psst post poij

In this kind of analysis, in external reconstruction, we try to discover cognate words from different languages by formulating regular phonological relationships among the words. Once we discover these relationships, we can be fairly certain that the words descended from some roughly common source. We can then reconstruct that hypothetical earlier form by piecing it together out of those sound features which correspond most frequently and most widely. No single entire word identical to an ancestral form need now exist-

520

GRAMMAR AND SOUND

The mosL widespread phonological form or feature (voicing, spirancy, nasality, for example) thai cannot be attributed to borrowing й a strong candidate for being closest to the original form, tn (1) in Problem 13.21, for example, the common segments in (1) could be reconstructed as /t/, jtl, /i/, or /MA (3,14, 91,12S, 188) PROBLEM 13.22: Here are some cognate words from several modern and ancient languages that we sball assume we already know are related. Reconstruct the phono) ogically earliest form for only the sounds in boldface. When you have sorted out the relationships, phrase them in this way: Sound X > sound Y. Then using the kinds of categories of sounds listed in Table 1ЭЛ, group the sound changes into categories of sound changes. Keep in mind that we are not assuming that any of the sounds in the words listed here came from any of the other listed words. Rather, all these sounds in boldface in all the words descended from some earlier unrecorded sound in some unrecorded word, a sound and word that may be identical to any of the relevant sounds or words or quite different. But assume that that earlier sound is probably similar to the more geographically widespread one or the one found in a much earlier langjage. The , , initial sounds in Sanskrit (the earliest records date from с 1500 в.с.) are voiced stops followed by a slight puff of breath. Unless otherwise stated, the non^Englisb word is Latin. The words arc spelled in ordinary orthography, not in phonemic or phonetic transcription. Give the boldfaced consonants their ordinary values, (th) spells cither /9/ or /5/, (Lith. = Lithuanian, Old Bulg. = Old Bulgarian, Sans. = Sanskrit, Gk, = Greek, Russ. = Russian, Pers. = Persian.) fliorp-troba (Lith. house) yoJsc-yugam (Sans.) bear (carry)-bhr. (Sans,) nephew-nepos deep-dubus (Lith.) f ive-pente (Gk.) fee-pecu (cattle) have-capere fish piscis lip-labia thunder-tonare к nee-gen u tbree-tres foot-pes flood-plotos (Gk.) apple abhul (Old Irish) cold-golotu (Old Slavonic-ice)

sleep-slabu (Old Bulg slack) tootb-dentis mead-madhu-(£aris.) hundred-centum brother-bbratr-(Sans.) two-dva (Russ.) father-pidar (Pers.) ten-decem do-dha (Sans.) acre-ager hemp-cannabis head-caput kin-genus heart-cridbe (Irish) cardhas (Sans, crowd) herd hump-kumpas (Old Slavonic) grind-ghfiihati (Sans.)

PHONOLOGICAL CHANGE

321

PROBLEM 13.23 ; If the change had not occurred, how might we now pronounce these next words? cakc,fb(, feed, fight, hawk, hunger, teach, throw, thing, town, true, heal, holy, finger, deed, deft, dare, gate, goose, ghost, comb, quick, dub, care, both, hope, hobby, pear, pine. PROBLEM 13.24: There seem to be in English a good many exceptions to these sound charges. We have not only tooth with /t/ derived from IE */d/,s but also from the same root, dentist, dental, indent, and so on with the unchanged /d/. We have heart with /h/ derived from IE */k/, but also the related cognates courage and cardiac with unchanged /k/. Why? The law which Problem J3.22 suggests was formulated by Jacob Grimm in 1822. But a number of nagging exceptions seemed to contradict his generalization. Where the rule predicted the IE /p/, /L/, and /k/ should become the voiceless spirants jfj, /в/, and /h/, they instead became Germanic voiced spirants, fb) (a hi labial spirant), /S/, and /y/ (a velar spirant), which in OE became fbj, /d/, and /g/. In addition, ;'s/ in some cases seemed to become /r/, a change apparently not related to GrirnnVs Law;

,

/t/ in IE * pater did not become /в/ but /d/ in OE fader (father), /p/ in IE *sept did not become jfj but /b/ in German sieben (seven), /k/ in IE *swkrus did not become /h/ but /g/ in Old High German jwigar (mother-in-law). /a/ in IE *wes- did not remain }sj but became /r/ in OE wsron (were).

In 1875, Karl Verner discovered that the reason for the discrepancy had been obscured by another, subsequent change. In (E, two syllable words could be stressed on the first or a subsequent syllable: IE *pater > (father) vs. IE *bhraater (brother). This difference disappeared in the Germanic languages when the stress in all words shifted to the root syllable: Thus IE *pater > Gmc *fa$er (the /t/ > /Й/ change was an intermediate stage before /di). Verner discovered that if a relevant consonant were In a syllable following an unstressed syllable, then Grimm's Law did not operate. It is for this reason we have pairs today like seethe jsoAAeu, wasjweie, raise jreur, losej (for)lvtn. In each case, the second word of the pair was originally a polysyllabic word stressed on what was earlier, a second syllable. The first word in each pair had initial stress. The same IE sounds, */s/ and */t/ became two different sounds, */s/ or */r/ and */S/ or */d/ because of that difference. 13.25: Assume again the year is 2550. You have just discovered in the ruins of a city once located on the southern tip of Lag Nizhian a chest 2. Here, we return to the us* of * to symbolize a root.

322

GRAMMAR AND SOUND

Full of boots,, letters, diaries, newspapers, and so on dating from about the last third of the twentieth century. Here are some excerpts from the materials. Assuming we can understand Lhem and many other documents we ha ve found, what kinds of tentative judgments can you make about pronunciation from them? 1,

A letter from a seven-year-old to a friend: Deer Jim, wel, it luks light them sientistshav dun it at last. Thei sed that they new how to fiks up (he woler in the leyk so that it wood nat meyk poyzin gass eny mor. They are goin to pump it ful of orinj jus. I hope they are rite becaus 1 shoor am tird of liwing undergrawnd like thiss. My father SEZ that it wasn't becauz of biz fakteries pcllutin the wander. Jt was a I the bippyz talc in bailis in it. The sientists say that evrythin will lie OK in just a fyu munths. I hope so. Yer frend in Ellinoy, Turn.

1,

A poem: I think that 1 shall never see / A bug as lovely as a flea, / A flea who hops around all day ;' And never has to ride a sleigh /A bug whose passion burns all night /And never bites you out of spite. / So every constellation's sun / should shine its brilliance on such one.

3,

Some jokes: я. When the three children of a rancher inherited his ranch, they called it " Focus " because that's where the sons raise meat. b. Why did the fly fly? Because the spider spied her. c. Knock knock. Who's there? William. William who? Will y^ meet me 'round the corner in half an hour? d. What's black and white and red all over? A newspaper.

4.

Advice from a spelling book: Do not Confuse the spelling of these words: ajffiect—effect, principle-principal, accept-exce-pt, emigratit-inirtiiffKin t, coinplsincnt-compk'ment.

What other books, documents, and so on, would you like to have? This kind of evidence is used for internal reconstruction. Linguists use spelling, particularly by those who are uneducated, or better yet, halfeducated. There are two kinds of spelling errors, however. First, simple misspellings. If someone writes dum for dumb, we know the
PHQNOLOGTCAL CHANGE

323

spelling and pronunciation. Ft is not always possible to specify exactly how a word was pronounced, but it is often possible to estimate whether two words were pronounced almost alike, which is almost as good- And if a good rhymer never rhymed two words that were spelled alike, we also know when two were probably not pronounced alike. PROBLEM 13.26: Here are some rhymes and puns from Shakespeare (c. 1600), Pope (c. 1700), and Wordsworth (c, 1800). Comment. Shakespeare; tears {iioun)-hairs, case-ease, eate-hate, say-sea, shapeskeep. Pope: weak-sake, eat-state, shade-dead, speak-take, sea-obey. Wordsworth; fiees-ptease, rei;ea!ed-stee!ed, heal-feel, peers-years, dreams-seems. PROBLEM 13.27: Here are a few dated spellings with OE and ModR related words. Comment on the consonant!;. /?еАл/-1400 (ОЕ Ы-heatfe, ModF, behalf); bight-1500 (OE bit an, ModE bite); faid-1450 (OE fid, ModE/он/), dvombe-159% (OE dom, ModE doom), wore-1350 (ОС. werre ModE war). Another, more complex way to reconstruct phonological history from internal, contemporary evidence is to examine how current sounds are distributed in the words of a language. For example, /v/ never occurs as »the initiai sound in ModE words derived from native OE words (with the exceptions of vat, vixen, vent (as in coat rent), and perhaps pane). Unless a sound change devoiced all initial /v/ sounds, we can assume thai all OE words had only inkial ffj sounds, PROBLEM 13.28 r Examine the linguistic sources of words beginning with /j/ £nd /z/, or which have fit anywhere in them. Comment. PROBLEM

13.29: What is unusual about the distribution of /rj/ in English?

It should be noted here that the data in the following Problems are rattier misleading. They arc neat, clear, and consistent. But in perhaps no other area of historical linguistics is it more difficult to interpret the data. Beginning with the mere decipherment of the often dim and illegible handwriting она crumbling manuscript, the problems of interpret ing the relationship between spelling and pronunciation abound. What were the conventions for representing iounds? Were only phonemic distinctions observed? What do vowel digraphs like or (ea) represent? Who was the original author of the manuscript? What was his dialect? What was his social class? How old was he? When did he write the manuscript? Did he writt it in his native dialect area4.' Or was he

324

GRAMMAR AND SOUND

influenced by another dialect area? How educated was he? Tf he was badly educated, how can we know whether misspellings were the result of attempts at phonetic writing or of i gnu ranee 7 Js the manuscript a copy? Who copied it? What was his dialect? His education? How old was he? Was the scribe a consistent speller? Were there historically two forms for a single word? In regard to poetry, did a particular poet rhyme accurately? Did he choose variant forms of words for rhyming purposes? Did those commenting on spelling and pronunciation really know what they were tailing about? There is some evidence Lo suggest that many did not. In short, there is very little in the following Problems to suggest the practical difficulties a historical linguist faces in the attempt to reconstruct from the best texts pronunciation that has long since disappeared. (214)

THE PHONOLOGY OF QUA ENGLISH

We have available several different kinds of evidence to help us reconstruct the phonological patterns of a dead language, but we must have a key to the. system to begin with. If we discovered a completely unknown languagewritten in a completely unknown orthography, no amount of purely written evidence would suggest how to pronounce it. At one time, scholars had no idea how LO read Egyptian hieroglyphics. But the discovery of the Rosetta Stone in 1799 with its simultaneous translations of the same passage in Greek. Egyptian demotic script, and Egyptian hieroglyphics gave them the key. Unfortunately, no such key iias been found for Lhe writings left by various South and Central American cultures, making them largely indecipherable. hortunately for those investigating the history of English, however, the Latin-based script used by Latin and Irish monks corresponded rather well with the spoken Latin on which it was baser], a language we do know how to pronounce. Thus given the Latin letter <[> and our knowledge that in Latin it represented only cither short /i/ or long /I/, we can conclude that the OE word written Urn was certainly not pronounced like the Modli descendant of the form, not like the ModE lime /lairn/, as in bird-lime.

Vewels The letters used for vowel sounds in Late OE (с. 900-1100} were these: <e>, <se>, , , , (u>, . Earlier manuscripts (c. 800) used as well, buL this letter disappeared before Late OE, We can assume that the

PHONOLOGICAL CHANG Q

325

sounds these letters represented roughly corresponded to their Latin equivalents: = /a/ or /а/ = /i/ or /i/

<ж> = /ж/ or ,'£/ - jo/ or /a/

<e> = /e/ or ffl = /u/ or /0/

(The letter we shall deal with in a moment.) PROBLEM 13.30: Unfortunately, Anglo-Saxon scribes did not use diacritical

murks (lines over letters, strokes, dots, and so on) to distinguish long vowels from short. Indeed, the lack of any such marks unambiguously indicating length means that we do not know from OE orthography alone whether longshort distinctions even existed in OE. Here, however, are some data that will suggest whether long-short distinctions did exist. You can use both internal and external methods of reconstruction, (Assume Lhat doubled letters and two-letter sequences like <ei>, (an), are diphthongal or long in the nonEnglish examples. LcngLh marks have- been supplied in some cases. OE lim > ModE Hmk-limr (Гее.), lern (Dan,, Swed.) OE lira > ModE Itme-iijtn (Dut.), lim (Ice.), Him (Dan.), Seitn (Cer,) OE hem > ModE home-keim (Dut.), heirttr (Ice.), heim (Gerr) OF, ham > ModE ham-hamms (Ger,), horn (Ice.), hat» (Dut,) OE dif > ModE diff-klif{D\xx., Ice,), klippe (Swed.) OE wij > ModE wifu-wijf (Dut.), tj/(Ice.), weib (Ger,) * OE ful > ModE/^/-™//(Dut.),/^{Ice.),/jw/(Dan.),/ i ju/(Ger,) OE/«/ > ModEfuJt-wl (Dut.),./i(№ (Ice.), fuli(Swed.), tw//(Ger.) OE d&d > ModE deed-daad (Dut., Dan.), dS8 {.lee.), deih (Gothic) OE bed > ModE bed-bed (DuL), best (Ger.), beidi (Gothic) OE hot > ModE boot-boete (Dut), hot (Ice.), boio (Gothic) OE fop > ModE lop-lop (Dut., Dan,), toppr (Ice.), sopp (Swed.) OE dmi > ModE deai-deel (Dut., Dan.), daSs (Gothic), theil (Germ.) OE st > ModE ui-ai (Ice, Gothic), wi/(Dan.) From the external evidence, from comparisons with otlier languages, we can conclude that OE had long and short voweis. The evidence shows that for the most part, cognate words in Germanic languages have corresponding long-shorl distinctions: Danish iem Sum. Since it is unlikely that this difference developed independently in Modern Danish, German, Swedish, Dutch, Icelandic and in Gothic, but not in OE, we can assume that OE also inherited length from its Germanic ancestor. On the basis of the internal evidence (though there are many exceptions to this generalization which we shall explore below), words which had long vowels in OE correspond to ModE words with diphthongs: Of.lim > ModE lime. Words with short vowels Ln OE have monophthongs in the descendant

326

41! ли мл к AND SOUND

ModE words: OE Um > ModE limb. Apparently then, ModE diphthongs arc the regular modern reflex of OE length. If we can trust these conclusions, then we assume these OE vowels: I/i

и/й о/о

£/e ffi/se

a/a The precise quality of any of these sounds is impossible to reproduce of course, but they must have been close to these: /i/ as in sit] ;'?/ as in a monophtbongal pronunciation of seal; /e/ as in set, /8/ as in monophthongal sate; /a;/ as in cat; /aS/ as in cad; /a,1 as in hot, /a/ as in hod\ /u/ as inpu!lt /й/ as in monophthongs I pcoi^ /a/ as in bought, jvj as in boat, PROBLEM 13.31: Because for a time, the letter was consistently used only in certain words where we now have ModEhide-huid{Dut.) OEfyr > ModE fire-vuur(DiiL) OEcyssan > ModE kiss-koss (Ice.) OE суп ing > ModE king-konung (Swed.) OEsynn > ModEsin-siindc{Ger.) OEbryd > ModEbride-Ьгпбг OE yfel > ModE eviMlbcl (Ctr.) OEfyst > ModEfist-vuist(Dut)

OE fyllan > M odE fill-fulljan {Goth.) OHdynnan > ModE din-don (Dan.)

The cognate languages have rounded vowels, front or back, where OE had . But in most ME texts, this came to be consistently res pel led , indicating that the sound it represented had by then merged with the high front /i/ and /I/. Therefore, in OE almost certainly represented a sound that originally shared both rounded and front qualities. It was a rounded iiigh front vowel, an unilauted vowel that approximated an /i/ with rounded lips. The umiauting resulted from the change in pre-historical OE which we have already described. The digraph in Early OE represented a rounded midfront vowel corresponding to .'e/. It derived from Gmc *jal: *dainjan > tfamon > demon (deem). Since for both (y) and <,ce> long and short reflexes occur in ModE, we can assume four more vowei sounds in Early OE: ]/i

u/u ё/е 6/о

u/fl о/о a/a

PHONOLOGICAL CHANUE

327

By 900, !aj had coalesced with /e/. There is no evidence of a significant jsf phoneme, cither stressed or unstressed in OE. FROBLLM 13.32: Suggest a way to determine when the rounded front vowels became unrounded, PROBLEM 13.33: From iht data in Problem 13.30, we can also see л great many other changes in the quality of vowels. One meaning of OE litn /Urn/ is now pronounced /[aim/; the other is limb from short OE /lim/. OE hus /hfls/ is now pronounced /haus/ (except in Scots English where the /I/ > /ai/, /п/ > /au/ change never occurred. Scots thus hfis hus for house, ее for eve, and so on). Plot the changes between OE long vowels and ModE diphthongs. The dialect ofOE we arc describing here, West Saxon, may also have had eight diphthongs, four Jong and four short. They were written , , and <eo>. By A.D, 900, had simplified to , and had merged with (eo>. There is considerable debate about the phonetic values of these segments, even whether there were short diphthongs. But on the basis of subsequent changes in these words, they were probably close to these values: <ea> = ДО; - /ia/; <eo> = /»/; = /b/, (25,28, 89, 111, [48, 175, 209)

Consonants

Describing the OE consonants is somewhat more difficult than describing vowels because the consonants were less consistently represented. The letters used to represent OE consonants were these: , ,

, , , , <3> (yogh), ф ) (thorn), <«> (eth), <s>, <m>, , , ,

(wynn), and . <x>, (z), and were rather rare. Modern <w>, it may be recalled, was represented as or

. Because the relationship between the following OE letters and ModE letters has been extremely stable over the years and because the ModE Setters still correspond to Latin phonetk values, we can assume that in OE, they represented sounds very close to the same sounds in ModE: , <^)> , <m>, , , and <w>. was probably different only in quality, very likely being more like the trilled /r/ in Modern Scots. PROBLEM 13.34: Transcribe these OE words. You will have to work backwards from ModE to determine vowel length: (bath), bttm (bite), си (cow), cwic (quick), fat (deal), dxd (deed), kus (house), deman (deem), dom (doom), rfwetttin (dwell), endian (end), fset (fat),/«*M (feed) Jij (five), fyth» №)> giadi$n$> ffM (goose), hal

32S

GRAMMAR AND SOUND

(whole), hattg (holy), hdpan (he\p)tfrogga (frog). Sand (land), lim (limb), fttf (louse), melton (melt), Jafla (bull),га*(sea), Шоп (sit), m t e (sweet), telim (tell), &w? (broth), pus (thus), r # (tide), /y« (thine), we (we), wfe (wise), gkes (glass), writan (write), god (God), ramm (ram). Some of tbe earliest changes in the inventory of English consonants involved the sounds represented by and <з> (represented by in ModE textsj recall, the symbol we sha.il use here). PROBLEM 13.35: Here arc some data from OE and some related cognate languages. (1) On the basis of the cognates, what probably was the earliest pronunciation of the OE digraph <sc>? Of ? (The sequence of letters <sh> and the sound /S/ did not occur in Early OE.) (2) Is there any way to predict which Oli words would come to have initial /s/, /к/, and /6/? OE iceacui > ModE shackle {skagle, Dan,) Otitffl/f > ModE shaft (skapl, Ice,) О E sceacan > Mc4E shake (skaka, Swed,) GEsceap > Mod shape {skapa, Swed.) OE cece > ModE cheek (kek, Swed.) OE corn > ModE corn (kaurn, Gothic) OE an > ModE chin {kind, Dan.)

OE serift > ModE shrift (skript, Ice.) OE scip > ModE ship (skib, Dan.)

OE se&h > M od E shy (skygg, Swed.) ОЕжея/ои > M.odB shave (shave, Dan.) OE ceaf > M od E chaff (kaf, Dutch) OE crapp > ModE crop (kroppr, Ice,) OE cttf > ModE diff{klippat Swed.) O£ en r > Mod E гаг (katze, Germ.) OE dele > ModE chHl (kola, Ice.) OE calf > ModE c.alf{kalf, Swed.) OE а и л > ModE chicken (kjuk!itigt Ice.) PROBLEM 13.36: If in OE, all sequences of [sk + front vowel] became [s + front vowel], how does it happen that we have /sk-/ sequences in ModE like skip, skim, sky, skin, schedule, scale, zkidt ski!!, skit? PROBLEM 13.37; Are any of these words apparent exceptions'^ Comment: keep, kettle, kelp, kitchen, kith, kin, kipper, kiss, kit, kink, chocolate, choke, coach, pooch, catch, pouch, touch, roach, brooch, hootch. PROBLEM (3,38: Roughly the same thing happened with the /g/ phoneme, though this is even more complicated because (that is, <з» represented

PHONOLOGICAL CHANCE

329

several different sounds. Here are several OE words with their modern descendants. C o m m e n t 1. 2. 3. 4, 5. 6. 7, &. 9. 10.

pegn-thane sixgen-slain twegen-twain ge-ye glsd-glad dzg-day falgsan-follvw i gast-ghosi

11. 12. 13. 14» 15. 16. 17. 18. 19. 21).

weg-way boga-how gear-year geoc-yeke geon—yon grag-grey a^a/HW/j haiig-hoiy fttglere-fowter gat-goat

21. 22. 23. 24. 25. 26. 27. 28. 29. 3U.

god-good gruel-gnat stigrap-stirrup menig-many gvld-gold grand—ground dragan-draw ssgdesaid gfowen-glow geotu-yelhw

The letter (g> represented at [east three sounds, but only two phonemes. One phoneme consisted of the allophones [g], the voiced velar stop, and [yj, the voiced velar spirant, two sounds which never occurred in the other's environment: [yl occurred only inside a word before or after a back vowel: lagu [!ayu] (law). Initially before back vowels or consonants, represented the other allophone, the [g] we hear today: gasl [gast] (ghost), god [god] (good). The third sound represented by the letter (g) was a voiced palatal spirant that was probably very close to a strongly aspirated [y],1 as it-might be stressed in year. It normally occurred before or after front vowels: sligrctp [stiyrap] (stirrup, literally was pronouneed [g]: singan Isirjgan] /singan/ (sing); bring [Jtring] /hring/ (ring). PROBLEM 13.39: This does not account for these words. Comment, get, give, gmkfi, guess, big, cog, dog, dig, Jig, hag, jig, tag, ieg, nag. The phonemic symbol /n/ used above in /singan/ and /hring/ should not be interpreted as [n]. Before the velar consonants /k/ or /g/, the allophonc of /ri/ was undoubtedly [rjj. the velar nasal we now hear in sing [sinj. If the allophone of /n/ before /g/ or /k/ was always predictably [rj]> then OE did 3 This is urJinarLly represented by Jj] i[) the International Phonetic Alphabet.

330

ОИАММЛИ AND SOUND

not have the three nasal phonemes we have: /m/, /n/, and /n/. It had only two: /N/ and /m/, with the allophones of /N/ being [n] and [n]. The combination written OE /ej/, thereby creating another phoneme, /J/. The letter represented in initial position the /h/ we use today' hecg [hej] /hej/ (hedge). But it also represented two sounds that we can hear today only in Scots and other northern British dialects: a voiceless palatal spirant after front vowels, symbolized [*]: nihi [ni*t] /niht/ (night), and a voiceless velar spirant after back, vowels, symbolized \x\: fukt [fuxt] /fuht/ (moist). They can be approximated by very strongly stressing the /h/ in huge and hoot, respectively. Because [h], [j], and [x] do not occur in one another's environments, they constitute the three allophones of /h/. PROBLEM 13,40: "We have already seen that initial /v/ and fzf sounds in English

words almost invariably indicate foreign borrowings, good evidence that OE probably had no voiced /v/ and /z/ initially. We can conclude that it had no Щ phoneme at all, since evidence for the sound does not occur until Early ModE. It also had no initial /j/ sound. Spirants that occurred finally in Oli were also voiceless. But it is clear that many voiced spirants in nativewords now do occur in final position. Here are some with their OF sources: love < lufii, bathe < bafiian, Нее < lifitm, dive < dyfany low < fasten. Comment, The OE spirants thus appear to have had voiced and voiceless allophones depending on their environment. In other words, instead of the set of voiced and voiceless phonemes as in ModE: ModE:

/f/

/6/ /s/ №

M № MЩ OE had only one sequence of spirants with two allophones each (with the елсерйоп of/s/, which did not have a corresponding /s/ and /h/). OE:

/F/ Щ

Щ

Щ

M

The voiceless allophones of the spirants occurred initially and finally and contiguous with voiceless consonants. Voiced allophones occurred when surrounded by voiced sounds. Thus the /F/ and /S! phonemes in ofer (over) and leosan (/аде) would have been pronounced [aver] and [leazan]; in of: (often) and dusi (dusr) as [eft] and [dust]. Given this pattern of voiced and voiceless allophones, it h virtually certain that OF also had only one interdental spirant phoneme that we have arbitrarily represented as /в/. It had two allophones, [S] and [6), distributed

PHONOLOQJCAL CHANGE

331

Jike tlie allophoncs of fFf and /S/. No native ModE words have a single voiceless /0j medially, and the voiced /Й/ sound now at the end of ModE words like bathe resulted from the loss of voiced inflections or analogy. Those words which now have initial voiced ,•'$/ are all roughly the same kind of words: the, this, ttuit, these, those, then, there, thus, though. They are all lightly stressed words whose original initial [G] in Lhe context of a whole sentence probably became voiced [$] in late ME. It is easier to keep the vocal cords vibrating in an environment of voicing than it is to stop voicing and then start again. It is the source of the difference between /9/ and /0/ in withstand and withail. (25, 28, 148, 152, 17S, 209) We can now contrast the overall system of consonants of OE with the overall system of ModE consonants. There are really three stages of OE consonant evolution: 1. 2. 3.

The development of /£/. The split of /k/ into /k/ and /c/. The split of the spirants into voiced and voiceless pairs.

Between Late OE and ModE, three more significant changes were yet to occur: 4. 5. 6.

The velar and palatal spirants represented by and would be lost. Щ would be acquired as a phoneme. /N/ would split into /n/ and /n/.

PROBLEM 13.41: Using a chart like Table 13.1, sort early and late OE consonants phonemes into points of articulation and manners of articulation. One last significant difference between OE and ModE consonants is that OE had nut only long and short vowels but long and short consonants. Long consonants were signalled by double letters: cysssn, moSSe, and they seem to Slave been always voiceless. They were probably pronounced as we pronounce the sequence of two identical consonants at the juncture of compound words. Compare these ModE pronunciations of phonetic doubled consonants with the pronunciation of words which have doubled letters only to mark the vowel quality of the preceding syllable. in koppian like the in ioppost. Compare in ebba like the in ribback. Compare in ModE rubber. in enytum like the in ttixd&e like the

in birddog. Compare

in ModE ladder. infrogga like the in doggun. Compare in offrian like the in Iffefoad. Compare <£f> in ModE offer, <S5> in ttxesse like the <ss> in busstop. Compare <ss> in ModE lesson.

332

ORAMMAR A^fC^ SOUND

(nn) in dyttnan like the <»n)in penknife. Compare in ModEspinner. <mm> in sH'imtnan like the (mm) in homemaker. Compare <mm> in ModE plummet, <S©> in invffie like the in wreaihikom. Compare
Chapter 14 PHONOLOGICAL CHANGE: FROM OLD ENGLISH TO MODERN ENGLISH

fHE MATURE OF SOUND CHANGE We have now described the two extremes in the history of English sound patterns: Modern English and Old English, The problem now is to recreate how the one changed to the other, when the changes occurred, and that most difficult of all questions, why they occurred. When we talk about how sounds change, though, we have to recognize the different ways in which they can change. In some cases, the changes are minor: Some words add or lose a sound without affecting the system of oppositions in the language as a whole: heafod > head, spinl > spindle, dragon > draw, ihimitde > thimble. In other cases, we change the ways we systematically combine sounds. In OE, for example, jka-j, /gn-/, /ha-/, /hi-/, and /hr-/ were acceptable sequences of consonants: cniht (knight), cnyUm (knit), gn*t (gnat), gnagan (gnaw), hnutu (nut), hnitu (nit), ЫфтА (lord), Meapon (leap), hnm (rime), bring (ring). Since then, /k/, /g/, and /h/ have been dropped from those positions, but not from the inventory of" sounds. A few new combinations of phonemes have entered the language through a handful of borrowed words: /pw-/ in pueblo, /sv-/ in svelte, /sf-/ in sphere, /ski-/ in sclerotic. But the minute number of words with these combinations has not affected the system ot 333

334

UK АММАН AND SOUND

combinations in English. Л( the end of words, on the other hand, some new combinations resulted when vowels between some consonants disappeared: glimpsed', once pronounced /gllmpwd/, became /glimpst/ when the vowel was syncopated, or dropped in an unstressed syllable. More basic changes create new sounds. When /S/ developed from /sk/, for example, the new spirant phoneme was a new sound that contrasted with other spirants. In other cases, sounds can change considerably without creating any new systematic contrasts. This occurred with OE /k/ before umlauted front vowels unrounded. The [c] altophone of ,'k/ developed before front vowels but did not systematically contrast with the [k] allophone of /k/ until joj unrounded to /еД And some sound classes can be lost from the inventory. This reduces the number of contrasts, as when the rounded long and short front vowels unrounded to ftf and /i/. This reduced the number of contrasting vowels from sixteen to twelve. But specific sounds can disappear without reducing the number of contrasting phonemes. This happened when some of the spirant allophoncs of /g/ disappeared, /g/ still contrasted with the other consonants, but in a more limited set of environments. Perhaps the most basic kind of change is when the system of features on which contrasts are based changes. Rounding is no longer a way to contrast vowels. Мог do we any longer use length to distinguish consonants. On the other hand, we have lost length in vuwcis, but the system of contrasts remains the same because we have substituted diphthongization. (J4, 85, 91, 115, 128, 188,209) Before we begin tracing any of these changes from OE to ModE, however, a warning is in order, Because we phrase a change in the form X > Y and because we have claimed that sounds change regularly, we might too easily assume that the history of sound changes is neatly linear, each discretely set off from the next, accomplished, as it were, almost overnight in all similar words, in a language with a single dialect in a classless society that speaks in only one style for all social contexts. Such is never the case. Sound changes are more like ripples m a pond, beginning in one place, spreading, merging with other ripples, interrupted here and there by irregularities in the bank. But in addition to geographical dialects, there are the ripples that spread through social classes, with pronunciations popular among one class spreading to another. And complicating this whole set of relationships is the fact that speakers from particular social classes within specific geographical dialects vary their pronunciation according to the social context they happen to find themselves in. In formal contexts they will speak differently than they do in casual contexts. Thus it is misleading to say that souxid X changed to sound Y in 1400 just because the first ncspelling of a particular vowel in a written text was found at that date. Sounds change first for someone in some words in some contexts in some social class in some geographical dialect. The change may

PHONOLOGICAL CHANGt; FROM OLD UNGLISII TO MODERN ENGLISH

335

spread through social classes and styles until it becomes the form used in the "standard" formal style in that dialect, or it may remain in the shadows of an isolated social or geographical dialect or style. Other dialects might remain unaffected by the change or borrow it. It might or might not finally affect Lhe core of the languaEe. We happen LO be describing one social -geographical dialect; American English spoken in relatively forma! contexts by educated upper-middle class speakers around the Great Lakes, There are other geographical and social dialects that we ctmld have examined. In some of these, certain of the changes we will describe below never occurred. So when we talk about sound changes, we have to specify four coordinates: location, social class, style, and period. Unfortunately, stylistic variations are difficult, indeed for the most part impossible to reconstruct. Differences among social classes are only a bit less difficult to account for. And it h not always clear how to determine (he precise boundaries of geographical subdialects. Specifying time, too, must be indefinite. We have to operate within a span of years: From the date of Lhc first citation which we might take as evidence that a sound has changed minus? number of years, to the date when the change seems to have permeated the formal style of the dialect being investigated. For example, the first indication that the palatal spirant [x] in words like niht was lost after front vowels seems to have appeared in mid-fifteenth century reverse spellings such as wnght for write. Yet otherwise reliable seventeenth-century orthoepists, dictionary-makers and so on could write a century and a balf later tbat a word Jik.e light was still pronounced by some conservative speakers with the spirant sound. Writing in 1632, Robert Sherwood commented that while he preferred the older pronunciation with the spirant, there was a more modern one used among Londoners, a pronunciation he seems to describe as /эу/. By 1650, the velar spirant seems to have disappeared entirely—in standard London English, But it is still heard today in parts of Scotland and elsewhere. So if we were writing a history of Scots English, this discussion about [x] would be beside the point. Describing a sound change i$ no simple matter if we are to be honest to our ignorance. With this in mind, we can explore some early changes.

OLD ENGLISH TO MIDDLE ENGLISH Vowels PROBLEM 14.1: In Early ME, in the twelfth century, words spelled with began to turn up in some non-Northern dialects spelled or . We now pronounce these words with /o/ as in boat and whole. By the thirteenth

336

GRAMMAR ANI) SOI IN Г)

century, many more words had changed to , but not all. Here are some data, What principle can you find to predict which vowels rounded and moved back and which did not ? O E b a t > ModE boa t-boot(Dut.) OEcat > ModE cat-katt (S wed.) OE hamrn > ModE ham-ham (Out.) OE swan > ModE swan-svanr (tee.) OEhai > ModE whole-hails (Gothic) OE salt > ModE salt-salt (SwetL) OEcalf> ModEcalf-kalv(Dan.)

OLban > ModE bone-been (Dan). OE ham > M o d t home-heim (Germ.) OE matin > ModE man-man (Swed.) OElam > ModEJoam-feem (Dut.) OEgat > ModE goat-geit (Dut.) OE hlaf > ModE loaf-laib(Ger.) OE ar > ModE oar-ааге (Dan.)

Dating this change of /a/ to /o/ is not difficult because both spelling and borrowed words can help. Since the womb, rtimban > climb, flatten > find, j eld > field, cumb > comb, did > childt seek! > shield, cynd > kind. In ME, land, hand, word, earth also became long, hut later shortened. Before /-mb/, /-nd/, /•Id/, /-rd/, and /-rO/, OE short vowels lengthened. But it was a permanent lengthening only for /i/ and /o/ before /-mb/: climb and comb; for /i/ and jit} before /-nd/: find-bound; and for all vowels before /-Id/: child, shield, bound, old On the other hand, long vowels shortened before consonant clusters other than the above, or before a stressed syllable: All the vowels marked long in these words shortened: softe, kepte, тёпе (dream), ftfiene, wtfman wisdom, ascian, bosnt, dust. So did unaccented long vowels in the second half of compounds or in unstressed prefixes and suffixes: arlsan, wlsdum, cynedom (kingdom), wurdllckc.

PHONOLOGICAL CHANOt: WtOM OLD tNGLISH In VUinhRN tNULiSH 337

PROBLEM 14.2: Here arc some data regarding some other words. All of these were short in OL. Comment, OE

ModE

OE

Modi:

OE

Modi

glsdlice

gladly паше apple spear step week skin lose god meat sun sack smoke

hUi-dre

bladder wood craft bath steal sit broth cup care weavel r»se mead staff

swingan

swing thus bathe

га ma appel spere slcppan wicu skinn losian god mete sunne saoc

smoai

wudu craft ЪзеС stelan sittan broS cuppe caru

wifoi rose med

staf

pus ba&ian ende bitela open

duru glses assa ofer tellan mcltan willa

end

beetle open door glass ass over tell melt will

Since many of the OE short vowels in these words have Ьесотле ModE diphthongs, those short vowels apparently lengihcncd. The environment in which short vowels lengthen is in a stressed open syllable optionally followed by only one more unstressed syllable. An open syllable is one which ends in ;i vowel: ri-tiat! (ride) as opposed to mel-lan. But as they lengthened, OE /i/, /e/, /u/, and ;'o/ also apparently dropped to the next lower sound: OE

ME u

a

Had they not dropped, we would have expected the lengthened /i/ and /u/ of wiku and duru to have become /i/ and /u/. And we know that long (if and /u/ eventually became /ai/ and /au/ sounds tid (/lid/) > tide (/taid.O, fats (,'hus/) > house (/haus/). But wikti becomes week, and duru gives deer (originnlLy /dur.'), the normal development of /ё/ and /o/, PROBLEM 14.3: So far, we have seen three important vowel changes before the end of M F : < 1) Lengthening before /-Id/, /-nib/, and so on; (2) lengthening in open syllables; (3) /a/ shifting back to a rounded vowel, /a/. Demonstrate

318

GRAMMAR AMD SOUND

the order in which these changes must have occurred. Use tin esc two words in your explanations: OL cold ,'kald/ (cold) and naitta /nama/ (name). The System of Contrasts When /a/ changed to A', it did not upset the system of ME vowel contrasts, however. It leu the language without a phonetic long [a], but the total number of contrasts remained the same: In Late OEand Early M H, we still find three long and three short low back vowels. But when short vowels lengthened before /-Id/, j-mbj, and so on, and in open syllables, ME regained an /a/ from the short /a/ in words like nanw, thereby creating one new contrast. There were then four long and three short low back vowels:

OE:

u/п ,>

о/о a /a

Early ME: u/п > after o/o

/a/ >№

a^

after

/a/

MF. : u/u o/o а/я

Unfortunately for the modern student, the /a/ and /5/ words were often spelled alike in M E texts, both with or : goose (f. OE gas), goot (f. OE gat), saoth (f. OE s6H)-hoom (f. OL ham). To know how to pronounce these words, one must know (1) whether the word goes back to an /a/ or an /6/ in OE, or (2) how the word is spelled in ModF. If the word is now spelled with spellings of certain <s) words in Late OE and Early ME when the ligature <sc) was still being used. Instead of usmg <5e> in al! words that originally had /ж/, some scribes began spelling these words with , presumably to reflect the new pronunciation of /a/. When short /it/ merged with a short /a/ and <se> as a symbol was dropped, words with ffl from OE were left with no distinctive letter to distinguish them from OE words that originally had /e/ or /a/. In the dialects we arc concerned with, Norman scribes consistently chose <e> or <ee) to represent words with an original /ж/, a fact which suggests that unlike short /se/, which had moved toward /a/, long /ж/ may have moved closer to /e/. As a consequence, words spelled <e) or <et> in ME were pronounced in two ways: /e/ if they derived from original /e/ words or from lengthened /i/

PHONOLOGICAL CHANCE: FROM OLD ENGUSH TO MODERN ENGLISH

339

words: OE sped /sped/ > ME spede /sp5d/ > ModE speed ;OE bite! fbiisilf > ME / K ? ^ /&#?// > ModE beetle. Or if they came from original /£/ words or lengthened /e/ words, (hey had a vowel that was by this time probably higher than /SB/ but still lower than /e/: OE J * [J^ л ] > ModE sea; OE wcfe ME fliefe [miE'tJ ModE meat Graphically, the sound spelling relationship resembled this: speed'Ispede (speed)

ё

о

<e> or <ee> ^ meet I mete

bSood]bhd (blood)

or

(meat)

яё

А

э

gootlgot (goat)

rtaamjname (a) or
The phonetic symbols usually used for the two higher front vowels are called opcn-e for the lower vowel that was originally {&} a ad close-e for the higher vowel opcn-e: [&], close-e l$]. (To distinguish them more clearly in this text, the more /se/-like sound, the open-c, will be represented with a large /£/ to suggest the wider jaw opening for lower sounds. The close-e will be represented with a small /5/ to suggest its smaller jaw opening-) Given these changes, the monophthonga) long and short vowels of ME looked like this: I/i

u/u ё/е Ё

о/б

a/a n.b. /o/ may have been closer to [a], PROBLEbf 14,4: Transcribe these ME words: boc (book), mild (mild), bac (back), bedde (bed), dene (clean), dele (deal), do (do), dwell(imtyjede (feed), /eld (fteld), fiod (flood), fate (foot), gold (gold), gras (grass), gost (ghost), hasppe (hasp), hyde (hide), AM (his), hog (hog), купле (kin), klyjfe (cliff), tokke (loek), m^/c (milk), nuwie (name), plate (plate), гйе (rise), saaf (safe), j/epe (sleep), ЙП/:С (sink). PROBLEM 14.5: What textual evidence would you look for to prove that two words spoiled or or <ee> were not pronounced alike? 14.6: How might you determine whether /ж/ merged with /a/ before or after vowels lengthened in open syllables?

340

ОНАММАЯ AND SOUND

In snme ME dialects, West Midland and in the Southwest, the long and short rounded front vowels from mutated back vowels, that is /S/, spelled spelling but an /i/ pronunciation: build, busy, business. Diphthongs

Between OB and МЫ, diphthongs underwent some important changes, so great that we cannot detail them all. Both OE diphthongs, long and short jsl and /eb/ disappeared, merging with long and short /ae/ and {if. But a multitude of new diphthongs entered the language, many as the result of a sound change involving velar and palatal voiced and voiceless spirants, allophones of OE /g/ and /h/. As we saw in Problem 13.38, the spirants spelled and eventually disappeared. We can date this ioss by noticing new spellings in ME and OE words that had these segments: ME dai for dsg, ME owen (owe) for OE agan. PROBLEM 14.7: Before or after stressed back vowels, the spirants were velar. Before or after stressed front vowels, they were palatal. Their loss resulted in this sequence of changes (represented phonemicaliy): 1. 2. 3. 4.

Palatal voiced: Oii dseg /dffiy/ > ME dai /daei/ > ModE day /dei/ Palatal voiceless: OEnea» Insshi > ME neigh /neih/ > ModEnig/r/nai/ Velar voiced: OE ЩП» /agan/ > ME owen /оиэп/ > ModE о we /ou/ Velar voiceless: OE phh /ploty > MEjp/ow/pluh/ > ModE plow /plau/

The voiced spirants, (1) and (3) disappeared first, leaving offglides in their places. Before the voiceless spirants, (2) and (4), ME vowels first developed an offglide, then much later the spirants disappeared. The offglide that developed in (I) and (2) was /i/ and in (3) and (4), /u/. Why did these particular ofiglides develop? Jn addition to these diphthongs, Norman French words borrowed into English contributed a few more, though most of these quickly merged with similar native diphthongs. The outstanding exception is found in rhois, noise, jute, and Joyn. The /oi/ pronunciation apparently changed to /ai/ later, since we find rimes tike join-fine in the eighteenth century. Our /di/ pronunciation is probably a result of the school master's insistence on spelling pronunciation. (25, 28, 98, 148, 150, 164,209,213)

PHONOLOGICAL CHANGE! FROM OLD ENGLISH TO MODERN ENGLISH

341

Consonants In describing some of The new dipti thongs in ME, we necessarily bad to describe one of the major consonant changes—the loss of the voiced velar and palatal spirant albphones of/g/. Because Norman scribes consistency res pel led OE words wilh the voiceless /h/ as while often dropping the (g) that represented the spirant allophones of /g/, we can conclude that the spirant ailophosies of /h/ were still present in ME. a conclusion reinforced by sixteenth century orthoepists who describe them. So far as can be determined, / i / did not appear in ME, We find no ME spellings of <sh> or (zh) for <s> in a word like visinn. So words which we now pronounce with /z/, as in pleasure, confusion, imasion, and so on were in ME probably pronounced /z/: [vizian], [pleziur], and so ou. And merely from spelling evidence, it is still impossible to determine whether the <-ing> in words like thing &nd sing represented [aigg] or [sir]]. As early as the fourteenth century, there are spellings of <-in) for (-ing) in words taking the past partieiple ending: haukin for hanking (hawking), but this has nothing to do with the problem of ,''-iij(g),' in non-participial endings, PROBLEM 14.8: If ME /N/ had not yet split into /n/ and /rj/, and /z/ was not in Lhe language, what did the phonemic consonant system aL the end of ME look like? (Use the kind of chart illustrated in Table 13.1.)

14.9: Here is a body of data that will suggest some of the common kinds of changes iu the way individual consonants changed in particular words to the end of the ME period. Comment. PROBLEM

OL leofman spinl pridda emtig hafoc hcaiod pymcl

ME ModE ;* 1cm man — ;• spindle > spindle ;» thirde > third ;> empti > empty > hawk ;* hank > head ;- bed > thimbcl > thimble

OE beorht byroen w ifman bridde gsers swa two

ME > briht > burdin > wimman > bird > grass > so > to

> bright > burden > woman > bird > grass > so > two

14.10: As we have pointed out, long vowels shortened and short vowels did not lengthen if they were followed by two consecutive consonants, or if they were followed by a stressed syllable. This has obscured the historical sources of some common Modli words. Had the vowels not changed, we would have had combinations like those below. To derive the ModE word (1) shorten the vowel and return it to its original Oli value and (2) unstress

342

GRAMMAR AND SOUND

the second syllabic, sheep + herd; moon + day; holy + day, bone + fire; cjear + -ling; goose + -ling; break + fast., olii{er) + man; roam + -Ые; wise + -ard, Chris! + nwis; house + bandit {bnnda meant \i freeholder, someone who owned his own house and land); house + thing tin the Danelaw, the ping was a council of elders. The descendant ModE word refers to political campaigning.}; good + spell (spell means story): nose + thyri (ttiyrl means hole); wind (as in wind up) + -I- + ass (ass means pole.); throat + -te {-te is an ending that shortens the preceding vowel.): wild + -er- + -ness; toad + poll (poll here means head; it is respeilcd рок in this word,); wife + man. (There is a secondary change in the first word. The initial /w/ causes the underlying shortened vowel to become rounded to /u/, and the disappears.)

MIDDLE! ENGLISH TO MODERN ENGLISH

We usually pick 1500 as the end of ME and the beginning of Early ModE because we are past the major grammatical changes: In nouns, the inflections have leveled to a plural and genitive; in verbs, generally to a third person singular -.T, a past, a perfect, und a progressive participle; and in adjectives, to a comparative and a superlative. Word order has settled into a S-V-(C) order, und prepositions have assumed their very large syntactic load. The auxiliary verbs continued to expand, particularly in combinations with the progressive. The verb endings -th and -esl were still active. And questions and negatives allowing either do- or non-
PHONOLOQICAL CHANGE: PROM OLD I NGLISII TO MODKRN ENGLISH

343

Vowels Bui in 1500, English phonology was close to the middle of a very major change in the quality of vowels, a change called The Great Vowel Shift In Problem 13.30, we saw that certain M E long vowels correspond with ModE diphthongs; ME

ModF

ModE

The earliest «spellings that indicate when long vowels began to change date perhaps from the thirteenth century in some dialects, so the change may begun as early as the 1200ns. /i/,> /el/and/a/ > /au/ Around the middle of the fifteenth century, there began to appear respellings like <ei> and <ey> for earlier representing /I/, along with respellings of <eu> and for early representing ;u/. This suggests that at least by then, /a/ and /I/ had begun to become diphthongs: hleytuifor btind, meyld for mi!d,feyr tor fire; faunde fotfoundt and south for south. The change probably began when an unstressed onglide was somehow introduced before ihe vowel. But then the strc^ shifted from the nucleus to the glide until the glide became the nucleus. The change probably went something tike this;

13«0

1400

1*00

1700

III > /if/ > /si/ > /ai/ /u/ > /uu/ > /iu/ > /au/ In 1540, William Lily claimed that the English pronounced the Latin Щ "too broad," which suggests thai they were by then pronouncing it either /si/ or /ai/. Since later orthoepists contrast some non-London pronunciations

344

GRAMMAR AND

as being even broader than that in London, it Is likely that in 1600, a word like rids was pronounced /raid/ in those dialects, /raid/ in London, /ё/ > /I/ and {Si > /0/ PROBLEM 14,11: Here are some spellings and dates, spidte for speehe (speech) (c. 1200), wypbtg For wepatg {weep} (c. 1300), spyde for spede {speed) (c. 1350), hy for he {he} (c. 1350), doun for done (done} (c. 1300), roude for rood {rood} (c. 1320), bloude for 6 Ы {blood} (c. Ш 0 ) , bone for boc (book) (c, 137O)r £ои<* for jfoorf (j?™^) (c, 1350). (Note: recall (hat was (he standard spelling for /п/, the standard spelling for /o/, or the standard spelling for III,) What does this suggest about OE /5/ and Such spellings as м/yping for wejwflg and bloude for Wcv/ indicate that as early as the thirteenth or fourteenth century in some dialects, /e/ had already begun to move toward—perhaps become—/I/, and /o/, to move towards— perhaps become—/u/. 14.12: What does this have to do with the problem of dating when III and /п/ began to diphthongize toward /ai/ and /au/7 J4.I3: Here are some words tha( had long /б/'s. Comment, moon, blood, good, spoon, flood', .S4wf, .mtwi. glove, shook, brood, must, cook, tooth, done, rook, stool, month, food, mother, look, goose, brother. /5/ > /o/ The lower long vowels also raised during late ME and Early ModE, but describing their development is more complicated. Recall that ME /5/ had two sources: (1) The /a/ > /5/ change between OR and ME, and (2) the lengthening of /o,' in open syllables: 1. 2.

Prom /a/ > J5j: stone, bone, boat, home, goat, rope. All these had an earlier /a/: stan, ban, bat, ham, gat, rap. From lengthened /o/: smoke, hope,fioat, robe, yoke.

In London English, this /a/ eventually moved up to replace the older /&/, which, as we have just seen, had fairly early raised to /u/. But in ME, recall that the original /6/ and ,'5/ were both spelled and spellings: btod and hoom. So spelling cannot help us determine exactly when /з,' raised to juj. By the sixteenth century, however, the lower joj had in many cases been consistently res pel led (oa) as in road (from OF. rod} to distinguish it from the /u/ that had moved up toward /u/. For the most part, this /u/ from earlier (oj was still spelled

PHONOLOGICAL CHANGE: FROM OlT) ENGLISH TO MODbftN ENGLISH

345

or , as mfaodjfode (Йтан/иЬ). From the descriptions of seventeenth century orthoepists, both pronunciations, /ft/ and faj for original ,'э/, seem to have been current, with the final selection of /o/ in Standard Elnglish coming sometime in the seventeenth century. /I/ > ;e/ and ffl > /i/ The most complicated problems of the Great Vowel Shift involve the raising of/a/ to /ё/, whose main source was lengthened /a/ in open syllables; and the raising of /Ё/ to /1/, whose main source was OF Щ and lengthened ME I&I- At first glance, the change seems perfectly straightforward:

1.

2.

/a/ > /ё/: /nama/ > /пйтэ/ > /nsem/ > /n£m/ > /пет/ (лвтг). Also included in this change were blame, same^jiame, dame, lame, make, rakey take, tale, some of which were original OF lengthened /a/, others from French /a/. /£/ > /i/ ; /sffi/ > /sE/ > /si/ (jra). Also included in this change were words like speak, mepl? meat, deal, heat, leap, read, №в/, steel, weak.

The dates of the /a/ > /e/ shift are difficult to pinpoint. Spellings of credyH for OF cradvl (cradie) in the fourteenth century and mede{n) for made in the thirtcenLh century suggest an early movement in ноте dialects. But as late as 1631, Alexander Giil condemned the pronunciation of ffl for wortk with as being faddishly uncouth. Name, according to him, should have ^ been pronounced something like /лЁт/, By the second half of the eighteenth century at the liiirst, however, the /Ё/ had raised to /e/ among educated Londoners:/nbm/ > /neni/.

/Ё/ > lit The ME /E,r words have been left until last because their hisLory is very confusing. As we- have seen, /£/ is between /e/ and /*/ phonetically, but probably closer to /ё/ since ME words with this appro*imate_sound were spelled <ee> or words, which had long since been raised to /[/: see j^Ufiee /fiT/. At about this time, however, we find lists of rhyming words and poetic rhymes that indicate London speakers were pronouncing these words with /Ё/ from O t /ж,1 (and from other sources) identical to words with original /a/ and /e/. That is,' we find in Shakespeare, for example, rhymes and puns based on pairs such as creafed-defeated, speak-break, great-defeat, nature-defeature, great-seat, jest-beast, sweat-heat, enlreats-frers, east-west, lease-excess, ennfess-decrease, appear-where, tsars-hairs, ear-hair, years-forbears, speafthere, fear-bear, were-appear.

346

GRAMMAR AND SOUND

From data such as these, it appears that at least by the late sixteenth century, /a/ words had merged with /Ё/ words, probably in a pronunciation very close ю (if. Bui then by the middle of the eighteenth century, this kind of rhyming and punning ceased. Instead, rhymes and puns began to [ink original jEj words: sea, meat, weak, beat, dear, with original /e/ words: see, meet, week., beet, deer. Since this /8/ had long since risen to /I/, the rhymes indicate that by 1750 those words spelled must also have risen past /ё/ to reach something very close to /i/ as well. Graphically the change in rhymes looked like this:

1700 h

seem Щ from M E /5/ /*Ю1И l&i from M E / £ / \

(seem /l/ from M E /e/ \ I800 {seam / i / from M E ,'E/J rhymes

мвк /ё/ from ME /a/

\sams /t/ from M E /a/ /

This would not have been unusual were it aot for the fact that those other words also pronounced fSj—those original /a/ words and some others— did not change to ,'J/ along with the / £ / words. Thai is, when m w /sem/ changed to /sim,'s the apparently honionymous sume jittni did not change. Nor did Name, dame,fame, came, lame, fame, late, fade, shade, cake, and so on. Large scale sound changes occur independently of etymology. The Englishman who al one time seems to have pronounced seam and name as rhymes in fif certainly had no idea they derived from different etymological sources and almost certainly never thought about their spelling as he pronounced them. But when he began pronouncing a word like seam not as 4cm.' but as /sTm,.1, he must have had some way to distinguish it from same Ihc apparently identical /sem/. Otherwise, he would have changed same to /sirn/ along with the originally homonymous seam. One theory is that Londoners were influenced by a non-London dialect in which /Ё/ had changed to /I/ much earlier than 1700, perhaps long before Londoners were even pronouncing /Ё/ as /e/. Schematically, it would have looked like this: 1250 London: Dialect X :

f&l ;• № !si

;

1550

1450

1350

1650

1750 ..-

>

/e/

;•

/i/—

J

>

1850

N

Only steak, break, yea, and great {along with several /e/ sounds before /r/: bear, pear, tear, and so on) seem to survive from that earlier London dialect. The problem, though, is to imagine how hundreds of thousands of Londoners, entirely on the basis of haphazard contact with some nonkital dialect, could change to /I/ only those words with /6/ that derived from OE jsbj or lengthened /c', and yet did not do what speakers almost inevitably do

PHONOLOGICAL CHANGE: FROM Ol-D ENGLISH ТП MODERN ENGUSK

347

under such circumstances—overgeneratizc. No words wilh etymological /a/, which in 1700 were rhymed with etymological ,•'£/ words, were mistakenly raised from /ё/ (о /1/. That is, we would expect that if reap and rape were homonyms, both pronounced close to ,'rcp/, then some few words like rape, tope, and take, might have mistakenly, by оvergcncriilNation, been raised to Щ, giving us reap, leap, teak, and so on. But apparently, this did not happen. We can perhaps елрЫп this if we rephrase the dialect theory in terms of social classes and styles. As we have seen, in each social dialect, speakers vary their pronunciation according to style. Pronunciation in formal styles differs from that in casual styles. When asked to read pairs of words, for example, upper-middle class New Yorkers will tend to pronounce a word like bod with a low front lie] much more often than with the sound that characterizes lowermiddle class speakers, a higher, fronter sound closer to [E] or even [е], making W a n d bad almost homonyms. But in more casual conversational styles, even upper-middle class speakers tend to use the higher vowels toward [E] in words lilt с bad more often. This argues that no geographical dialect is a pure, monolithic dialect with no predictable variations across classes or their styles of speaking. We have no reason to assume that the verbal behavior of upper-middle class Londoners in 1700 was any different. In their most casual moments, they may very likely have distinguished reap from rape words, pronouncing reap as either jript or closer to /rip/, while pronouncing rape only /rep/. When the orthoepists, themselves educated but not always upper-middle class speakers, made their lists of homonyms, they were undoubtedly describing their most л careful pronunciations, their most formal and prestigious forms: only /rep/ for reap. Since they could list only one form of a word, they could not thereby reveal the statistical probability that they would occasionally, even in formal moments perhaps, also use the non-prestigious form. Thus in с 1700 London English, Londoners may have been able to raise only certain fif words to /I/ while leaving behind other ,'c/ words: White they pronounced two different sets of words alike in a formal style, they pronounced them differently in a more casual style As lower middle-class speakers moved into the middle classes, bringing their pronunciation with them, they tended to use their natural ,'i/ forms. Since up per-middle class speakers may have had the same distinctions, though on a statistically smaller spread, they were able to distinguish the stigmatized form too. Thus when / £ / moved past /c/ to /1/ in the eighteenth « n l u r y , it was not so much a sound change in process us a casual form statistically expanding in upper-middle class speech, until it became the only acceptable pronunciation, even in the most formal contexts. The change itself must have occurred in casual speech much earlier, before OE /a/ and Isi rose to merge in Early ModE /S/. Perhaps the moral of this development is first, that pronunciations change —inevitably, and second that a new pronunciation will often be stigmatized. It certainly happened wiihihe/a/ > )ё/ change and with the Щ > /ai/ change

348

GRAMMAR AND SOUND

And third, regardless of the siigmalination, the new sound may eventually be accepted by upper-middle class educated speakers. We should not take this as license to accept or encourage any pronunciation lhal differs from the one considered "correct" by those who make their living Idling other people how to speak. There are inevitable social penalties to pay fur using stigmatized pronunciations in certain contexts. But the history of stigmatized sounds shouid caul ion JS not lo become outraged at what some seem lo feel is moral turpitude in "vulgar"' pronunciations. The fact is that, not always, but often enough, the once "vulgar" pronunciation has become the standard, to be as staunchly defended as it was once condemned. Short Among the short vowels since 1500, only the letters and (e>have more or less consistently represented their vowels. In fact, these two segments have been relatively stable since OE, Л few changes have occurred in individual words: Among many speakers, /i/ dropped lo /c/; since is often pronounced sence, /Я/as telt. Drydcn rhymed мим and prince, for example. But any word spelled with an in ME and Harty ModF that we know is short was pronounced /i/ or /e/, and if it did not lengthen, it is very likely pronounced that way today. The most confused and various relationships exist among the short lower vowels represented by and disappeared in Rarly M F when short /ж/ merged with short /a/, giving only two low or low Ьцск vowels: /a/ and /o/. But a number of further changes upset this relationship. First, since we do have words today with /a?/: cat, sag, hug/к and so on, sometime between ME and the present, English acquired a. short /a?;' again. And in the dialect area around the Great Lakes, at any rale, mosl words spelled with , suggesting an earlier /э/ or /o/ pronunciation, are now pronounced /a/: hot /hal/, knock /&ak/, top /tap/. Sometime after ME, Ihen, /a/ or /o/ unrounded to/a/. Dating the change from /a/ to /ж/ is extremely difficult because the spelling of /a/ and jsi is in both cases . It is at least likely lhat the /a/ > /ж/ shift occurred considerably earlier, probably during the sixteenth century. PROBLEM 14.14: Comment on your pronunciation of these words with were pronounced /a/, wart, waddle, swamp, swan, wax, quack, was, Jivfljj, quart, warn, dwarf, swam, wharf, wasp, squalid, wander, watch, swagger, wad, hat, santiffat!t qualm, craft, harm, farm, mil, war,fairer, quad, alley, tatiey, cart, squat Jailer, pad, path, has, warp, talk, walk,palm.

PHONOLOGICAL CHANGE! FROM OLD ENGLfSH TO MODERN RNGLHH

349

A nole of contemporary relevance here is that sometime in the seventeenth century, perhaps earlier, one dialect of British English lengthened [ft] to [ж] and retracted it to [a], as in father. \l is ihe vowci lhat we hear in British Received Pronunciation in words like path, glass, calf, dance, command, and so o r — i n general (though not invariably) before voiceless spirants and /n/. We ca n often identify an American speaker trying to sound like an Englishman when he self-consciously but incorrectly pronounces [a] in words where ihe [a] doesn'i belong, in words like t:ai, pack, cram, and so on. the typical behavior of a speaker emulating an unnatural paLlcrn of speechIn Eastern New England, [a] in alt the right places is the natural pronunciation, because the earliest colonialists pruhably came from the dialect area where this change hud already occurred. They brought it with them and established it in, among other places, Boston, where because of the social prestige of that city, the broad-a became the mark of education and politeness. Boston's close contacts with British pronunciation helped maintain the vowel.

014) PROBLEM 14,15: A change roughly similar to /u/ > /ж/ is illustrated by the spelling and pronunciation of words like hot, hng* rock, moth, nod, song, momy toss, rob, scoffs sop, soft, and doll. Assume an earlier pronunciation of the letter in all these words was approximately /э/. Comment. FROBLLH i 4 . l d : Did /a/ > /as/ occur before or after /я/ > /a/7 What does the spelling of slap for stop by Queen Elizabeth I suggest? 4

Between M E and ModE, then, the Great Vowel Shift and some changes in the shon vowels resulted in a somewhat different set of vowel phonemes, different enough, perhaps, to say lhat Chaucer's pronunciation of English words was closer to that of King Alfred's than Queen Elizabeth's. Yet on a deeper level, the changes are less radical than they might appear. Disregarding rhe multitude of M E diphihongs, the number of contrasts in the basic vowel system since M L has decreased hy only two. East Midland had six short vowels; ModE as it is spoken in most (but not all) parls of this country, has Seven.

ME:

/i/Je/,

ModE: lil.ttf.

,'u/, and/a/.

East Midland M E had seven long vowels. M o d i : has four diphthongs closely corresponding to ihcm, six if we count fm) and /au,.' from ME /i/ and /u/r

ME:

/t/, ItlJEIJafJaf.fGU and /Q/.

l

\S

ModE: /ai/, /i/,

I \ i

i

/e/, /о/, /п/, and /au/.

350

(iKAMMAEt ЛМО SOUND

Thus even though the pronunciation of particular words has changed, the system of contrasts has thanged much less. (43, 107, 108, 109 209 213 239 244) ' ' ' '

Consonant.';

Between the end of ME and the present, several changes have occurred in the distribution of consonants, but most involve individual sounds in individual words. PROBLEM 14.17: Here are some data. Are there any generalizations to be made ? What information about the data would you like, in addition to the dates already provided ? 1. 2. 3. 4. 5. 6. 7. S. % 10. 11. 12. 13. 14. 15. 16. 17. IS. 19. 20. 21.

laffe (laugh) 15(53 al f (half) 1389 behaf (behalf) 1442 wosted (worsted) 1450 wussfauppc (worship) 1480 morgage (mortgage) J448 offen (often) 1590 wich (which) 1494 when t (went) 1550 larneskynnes (lambskines) 1450 clyme (dimb) 1580 troff (trough) 1553 conschens (conscience) 1469 sawgeais (soldiers) 1550 owsold (household] 1451 sepukyr (sepulcher) 1450 Fakonbrigc (Falkonbridge) 1465 passelI (parcel) 1480 marster (master) I4S0 Chrismas (Christmas) J 639 wyped (whipped) 1550

22. 23. 24, 25, 26. 27. 28. 19. 30. 31. 32.

33. 34. 35. 36. 37. 38. 39. 40.

whear(wear) 1577 Lamhyth (Lambeth) 1418 dafter (daughter) 1634 thorf (through) 1465 sesschyonys (sessions) 1450 teges (tediojs) 1647 drynkyn (drinking) J3S9 pleshar (pleasure) 1642 craftyr (hereafter) J460 ihudd (should) 1531 (A lack of /t/ sound noted in these words (1701)): gristle, costly, bristle, ghastly, whistle, ghostly, glisten, fasten, lastly. blyn (blind) 1389 husbon (husband) 1450 colfcyld) 1550 baptis (baptist) 1389 нех (next) 1450 prompe (prompt) 1545 currup (corrupt) 1701 stric (strict) 1701

The last significant changes in the consonant system involve the loss of the velar and palatal spirant a I [o phones of /h/, the split of /N/ into /n/ and Ш* and the development of ffi. Some orthoepists in the 1600's were still claiming that the spirants in words like right, mgkt^fight, daughter, and so on could still be heard as very light aspiration. But rhymes, puns, and quibbles contradict their statements in some dialects. In Shakespeare's Venus and Adonis (1593), for example, we find these rhymes: white-downright, nigh-eye.

PHONOLOGICAL CHANOE: FROM OLD ENGLISH TO MODERN ENGLISH

351

\ght-despite, light-delight (an incorrect reverse spelling for deltle}, spite-light. The voiceless spirant was very likely lost to Standard London English at least by the end of the sixteenth century, in some dialects almost certainly during the fifteenth century. The split of /N/ into /n/ and /rj/ through the loss of /g/ after the fo] albphone of /N/ ([sirjg] > [sin.)) probably occurred sometime in the seventeenth century. At least it was not until then that the lack of a /g/ was specifically noted. The completely new phoneme /z/ was acquired in a few borrowed French words with /z/: rouge, genre, and so on. But in these words it has always been a foreign sound that almost inevitably was merged with /J/, as garage has become /gsraj/ from /gani?/. \z] has developed as a thoroughly assimilated native sound in borrowings like н ш , pleasure, osier, and so on in quite another way. When we rapidly pronounce combinations of words like did you, hit you, press you, or raise you, the sequences !-d + -y/, f-t + y-/, /-s + y-/, and j-z + y-j coalesce to become new sounds: n

j-t + y-j jej as in /hicu/ (hit you) /-d + y-/ HI as in /diju/ (did you) /-s -Ь y-/ /s/ as in jpreluj (press you) j-z + y-j !±l as in /rezu/ (raise you) J h e alveolar consonants /t/, /d/, /s/, and /z/ are pulled back by the following /y/ to become palatals- The spirants remain spirants bat the slops become affricates. Regardless of what those who prescribe "correct" pronunciation have to say about these patterns, they occur in the informal speech of even the most educated, Tn fact, they have resulted in the (if, /J/, /£/, and /z/ phonemes in the middle of many words. Such words as picture, soldier, issue, and vision were all originally pronounced /piktyur/, /soidyw/, /^уп/ and /vLzyan/. The evidence for these pronunciations comes from contemporary French pronunciations of such words, from rhymes, and from metrical patterns. A word lit condlciotm (condition) in ME was certainly a tour-syllabic word if lines these from Chaucer's Canterbury TaJei arc to scan correctly with ten syllables each: Me tkynketk it accordsamt to resem I To telk yaw all the condicioun I Ofech of hem By the Lime of Shakespeare, poets often intended spellings of -tior or -cioun to represent one or two syllable* as the meter required. Spellings likepi€churetsojertitshuziaU9O on began to appear ! n the fifteenth century, evidence that the phonetic change had probably occurred in the casual speech of some dialects considerably earlier. Spellings '^e t:ishiof] for vision, however, did not occur until the seventeenth century. «rhaps the change lagged su long after the others because the /z/ sound was

,Ш

ОНАММЛК AND SOUND

so completely foreign to English. The phonemes /£7, jj/, and /5/, on the other hand, were native sounds that occurred in other contexts: ship, chip, edge. A sound change that has had very extensive social consequences is the loss of /r/ in a variety of phonological contexts. The earliest evidence for the loss may date from the OE period, though it became a. much more common change from the fourteenth through the sixteenth centuries. Indeed, the loss of /r/ both before consonants, as in /ban/ for /barn/, and finally, as in /ka/ for /kar/ almost certainly occurred before this country was colonized, allowing those first colonists to bring an г-less pronunciation with them from their Southern and East Midland dialect areas. By the end of the seventeenth century, then, the inventory of phonemes and possible- ways of combining them into sequences was settled in Standard English. A few new combinations have entered the language through borrowings, but words like svelte, pueblo, sclerotic, and sphere are not part of any overall pattern of word formation. Among the consonants, we have since early OE gained seven new phonemes, most by a redistribution of old allophones in new environments; two, /s/ and /z/, because new sounds evolved out of old combinations. It is more difficult to compare the vowels because the shift from length to diphthongization means that we are dealing with two different systems. It is safe to say, however, that English has added a short /э/ to the short OE vowels, and has lost three long vowels, /*/, /a/, and /5/. (43) Here are three passages, one from OE, one from ME, and one from early ModE transcribed phonetically. The first is from the Buttle of Brunnanburg (translated on p, 60). It dates from about A.D. 937. Some of the inflections have been regularized, Hettcnd u-ungon Seotta Ieode / and scipflotan, fege feoI!onh,.' leid dennode fie\s sccga uwatc, / sLj?]?an sumie upp on irmrgentid, / msere tungol, glad ofer grundas, / Godes candel beorht, eces Drihtnes, / oS sio a;Scle gesceaft sah to setle.

hettend krurigsn sksM a ]ftjdc/ and li pftota n fc&LlanI feld dennade seja swatt / siSSan sunne up an тэгуеп cid / тяге tuijgal glad sver gmndas / gades kandcl bearxt eces dristnes /ов SHK ta setle

The second consists of the first 12 lines of the General Prologue to Chaucer's Canterbury Tales, written in the late fourteenth century: Whan that Aprille with his shoures soote [hwan flat april wi9 his Sures sCts The droghte of March hath pcrccd to the roott бэ dfUKt of marc ha9 perssd tO Уэ ruts

PHONOLOGICAL CHANGE: FROM OLD ENGLISH TO MODERN ENULLSII

And bathed every veyfle in swich licour and bao'sd cvri vxin in swit likur Of which vcrlTi engendrcd is the flour; af hwii: vcrtiu enjendrad is Ээ (Юг Whan Zephirus eek with his sweete hreeth hwan zefirus ek wi0 his swets ЬгЁ9 Inspired hath in every holt and heeth inspired haft in evri halt and hEO The iendre croppes. and the yongc sonne Й5 tendrs krappes and вэ yuijga sunna Halh m the Ram his halv[e] cours yronne, hafl in вэ ram his halvs кппз irunna And smale foweles ma ken melodye, and smab fOJes maksn mciadis That slepen al the nyght with open ye Sal slepsn al Оэ nixt wit* эрэп Ъ So priketh hem nature in hir coi'ages se prik^O hem natiur in hir kurajss Thanne longcn folk to goon on pilgrimage[s] , , , flan iongen folk to gon &n pilgrimaps . . . ] The third is Macbeth's well-known speech in V,vr: *

Tomorrow, and tomorrow, and tomorrow, [l-этагд a;nd tsinoro ^nd 1этэго Creepes in this pclty pace from day lo day trips in ois peti pEs frum dE ta dE To the last syllable of recorded time. ts Sa Jaest silabsl sv rskardsd tsim АпЙ all our yesterdayes have lighted fooles ffind al aur yestsrdC?, l&v I sited fulz The way to dusty death. Out, out, brecfc candle! 3a w£ гэ dusti deQ sut sut hiif ksndal Life's Ьщ a walking shadow, a poorc player Isifs but s walkirj ssedo a pur plEar That struts and frets his houre upon the stage Ext struts snd frcls his эиг эрэп 6э StEj And then is heard no more: it is a tale snd йсп iz h£rd no mor it iz э tbi Toid by an ideot, i'utt of sound and fury, told bai an \йЫ fui flv ssund aaid Signifying nothing, signify in

353

354

GRAMMAR AND SOUND

SOMt POSSIBLE CAUSES OK SOL'ND The question which most interests us, naturally, is why these sounds should have changed at all. Л few decades ago linguists generally gave up searching for specific reasons in the external history of a language realms like dim a tic changes, massive incidences of speech defects, diet, and the like, and instead, concentrated on describing how they changed. In one sense, of course, an accurate description of how they changed itself explains why they changed. Some phonemes split into two because their environments changed. This only pushes the question one step back to why the environment changed. We must look into the speaker for the cause of those changes. Most recent attempts to explain phonological change approach the problem from one of two directions: physiologically or psychologically. The least specific kind of physiological explanation depends on the idea of "phonetic drift." When we pronounce a sound, we can think of ourselves as aiming at an ideal point of articulation. Statistically speaking, however, we hit that exact point only rarely. Rather, like someone aiming at a builseyc. we scatter hits around the mark. It is through sheer statistical chance that the center of the spread defines the phonetic ideal. But also through statistical chance, the center of the spread can drift until, for example, an /a/ drifts to /se/, an /I/ or /п/ begins to drift toward diphthong, an /a/ drifts toward /5/. Working against this change is the presumption that sounds bearing a heavy functional load, sounds that are used frequently like /i/ or /e/, must not merge, lest too many words become confusing homonyms. Apparently, however, the increase in homonyms when /ж/ merged with /a/ in ME did not confuse speakers enough to prevent their merger. For many speakers in sevcntcenlh-eentury London Rnglish, the sounds that developed out of /£/ and /a/ coalesced in /ё/ (name-reap), while in other dialects they did not. an outcome that casts some doubt on whether functional load satisfactorily explains any of these changes. The opposite of one sound perhaps "pushing" another is one sound " pulling" another. If, for example, /I/ began to diphthongize as the first step of the Great Vowel Shift, to shift away from (if, then /e,' had morearticulatory " r o o m " to vary in, and so drifted toward the "empty" spot inthearticulatory pattern left by ;i/. When IE; began to drift up and forward toward /f,', a speaker was thereby allowed to spread bis /a:,1 " h i t s " higher and more forward without confusing ihe resulting sounds with one that had moved on. And as fsi drifted up toward /ЁД /i/ had articulatory room to vary in. (106) Thus we can talk about one sound pulling another or pushing another in a way not too different from the way we talked about one lexica! item in a

PHONOLOGICAL CKANCt: FROM OLD liNOUSH TO MODfRM F.NGLISH

3SS

pattern pushing another or leaving a vacuum and thereby pulling another item to replace it. Bui again, none of this «plains why any of these sounds began pulling or pushing in the (irsl place. One physiological way to explain some kinds of phonetic movement is to look hack at the way sounds influence surrounding sounds. Over and over, we have seen how one sound assimilates to another. As we pronounce words, our tongue moves toward one point in the mouth, hut our nervous system has already prepared itself to (ire off another impulse for the next sound. In some cases, the impulses pile up and jumble the articulation, as when we try to repeat a torgite twister like rubber baby buggy bumpers or she sells sea shells by the seashore. Umlaut ing is a case of anticipatory a ssi mi iat ion; so is [n] before /g/ pnd /k/. The velar and palatal voiced spirants in dscg and dragan eventually assimilated completely to the vowels around them, giving us day and draw. Thus one powerful force for change in the language is to be found in the neurophysiology of articulating sequences of sounds. In other cases, the sequence of sounds may be too similar and the Opposite of assimilation occurs. Sounds dissimilatc. This is less frequent than assimilation, but it happens fairly uflen: library becomes Ubary, athletics becomes atheletics. In talking about communicative efficiency and impulses in the nervous system, we have already touched on the psychology of sound changes. But motor skills and perception arc not as complex a capacity as learning ability and thut abstraction peculiar to man—linguistic competence. And once we move into these areas, we are in very uncertain territory. We can X-ray the tongue and mouth, but we can't X-ray the mind. One purely mentalistic way to explain sound changes requires us to understand the psychology of the child. When a child first learns a language, he usually overgeneralizes the rules. Instead of learning first once and for all that some past tense verbs change their forms: go-went-gone, rtm-rau-run, and so on, he reaches a point where he generalizes that all past lenses take -id: goedt mimed, eated, and then often overgeneralizes further to wen ted, rawed, and aled. It may be that some of the same overgeneralizations occur when they Icam sound patterns. For example, all of us have implicitly learned that in English only certain initial sequences of consonants are possible, others impossible: No spirant except /s/ may occur before stops; stops may occur only before vocalic elements like ,,'r,', /I/, /y/, and /w/; these vocalic segments m a y occur only before vowels. OE and ME, on the other hand, allowed these: /kn-/ as in cnihu fen-/ as in gnstt, /hi-/ as in Ыеарал, /hn-/ as in bnutu, /hr-/ as in hrlng. By eliminating such combinations, the general rules for combining sounds become simpler. If rules arc acquired for sounds as they arc for grammar, children may tend to simplify phonological rules as they barn them. In other words, language can change slightly as it is imperfectly

356

GRAMMAR AND SOUND

passed on from generation to generation, as each generation simplifies the rules of the language. (239) Another psychological explanation of sound change also depends on exceptions to rules, but here we might examine [he question more graphically. Once a pattern begins to develop among the phonemic oppositions in a language, it is possible for "holes" to develop in the pattern. If we look at the spirants of OE, for example, we have the one set:

that split into two:

www When the velar and palatal allophones of /h/ disappeared, /h/ more properly became a semi-vowel. This left only ill without a voiced opposition. It may be that the pressure of this pattern to fill out the systematic oppositions between voiced and voiceless segments allowed jij to enter the language more easily than it might have otherwise, thereby providing an opposition to fif, A quite different kind of psychological pressure toward linguistic change has just recently been investigated quite intensively—the pressure of social identity. Enough evidence has been gathered in recent years to suggest that a social group can adopt a new pronunciation that will linguistically identify itself in one of two ways: (I) If they want to isolate themselves from other groups, (hey can develop their own pattern among themselves, or (2) if they want to join another group, they can identify a pattern in that group and emulate it. Neither explanation, however, is entirely without its problems as a way to explain large scale sounds changes. The first may apply only to relatively cohesive groups, such as Black young people in the inner city or fishermen on Martha's Vineyard who have come to identify themselves as the last of the real live Yankees. Both have developed speech patterns significantly different from those groups they have isolated themselves from. The second explains why lower-middle class speakers overcorrect themselves and in formal styles use features of speech characteristic of uppermiddie class speakers more than those speakers do themselves. But it does not explain why so often in ihe past, a pronunciation stigmatized as lower class, faddish, uncouth, or dialectal, ultimately displaces the poh'le o r e : /ae/ for /a/, /a/ for /a/, /i/ for /с/, /ё/ for /Ё/, /ai/ for /ai,1 have all been criticized as improper or unfashionable. Yet at least for educated speakers around the Great Lakes eaeh has become the standard pronunciation. It may be that as lower class speakers rise into the middle class, they correctly identify and successfully imitate certain middle-class pronunciations. But they also bring with them other phonological patterns that they failed to

PHONOLOGICAL CHANGE: FROM OLD BNOUSH TO MOJHKN FNOUSH

357

recognize arc r o t part of up per-middle class speech. Thus while they successfully emulate up per-mid die class speech patterns in some respects, tliey overlook other characteristics o f their pronuncimion not part o f middle class speech. Only when enough ex-lower middle class speakers become part of the middle class to threaten an accepted pronuneialion do the onhtiepisis and handbook writers concern themselves with LIT encroaching " vulgarism." Ordinarily, they do not bother to condemn characteristics of lower class speakers that pose no threat to "refined" English. Few, if any, of the orthoepists of the sixteenth and seventeenth centuries, for example, condemned {{j for /Й/, mouffot нтян'Л. ол/ for vath, and so on, even though it was current among some speakers, for it never threatened to become part of standard middle class verbal behavior. (118, J21, 2Щ In the same way, few handbooks of pronunciation ever comment on /b/ for /v/ in debii for detftl, herу for very, and so on, because it is a characteristic of only a few lower class Black speakers. Such speakers pose no linguistic threat to upper-middle class speech patterns, so hardly any pre-1560 speechleaching or pronunciation manuals bothered to comment on i t

A NOTE ON AMKRFCAN PRONLl\
Had sounds not changed, of course, many of the differences between dialect areas and soda! classes in this country would not exist. For many years, scholars believed that American dialects began to differentiate after the main colonial settlement of the east. They thought that in the eighteenth century, Eastern New Englanders and Tidewater Southerners began to cbange the way they pronounced words because they were so receptive to British English influences, while those further inland who were more isolated did not change. Thus when educated standard English in London lost its {if in words like barn, car, and so on, the prestige of everything English made the Bostonian and Clwleslonian drop their {if too. When {of remained as the prestigious form in stop, cod, and fat, the New Lnglandcr developed his dislinctive fof in those words. When the standard British pronunciation of words like calf changed from [a?] to [a], the Anglophilie New Engender and Tidewater Southerner properly assumed iheir " broad-ah." It is the reverse of Bede's conclusion in regard Co OE dialects (Chapter Three). Encouraging this mimicry of British pronunciation, scholars also believed, was the schoolmaster, whose influence throughout New Lngland in particular was very strong. His usually lower-middle class background in combination with the general belief that teaching was a way to middle class respectability made him linguistically insecure enough to look to a more prestigious dialect

358

СЖАЧМЛН * N D S4JUMD

than his own as a model. And as a schoolmaster, he was able to drill his charges in those broad-ah*s and rounded-o's. More recently, however, some scholars have come to believe that American dialects differ because tiie dialects of the original settlers differed, because they came from different parts of Great Britain with their own characteristic speech patterns. In order to decide between these two views, we have to know from where in England the earliest American settlers came so that we can match dialect features that characterized those areas with dialect features of the settlement areas here. We must then reconstruct how the settlers migrated West, taking their speech patterns with them. Eastern New England and the Virginia TidewaLer were settled by East Midlanders and Somheasterners, the- New Jersey and Delaware area by a mixture of English, but many from the North and West. They later migrated east into central and southern Pennsylvania, Ohio, Indiana, and Illinois, and south into the Piedmont. Scotch-Jrish from Ulster settled western New England, upper New York State, and along the Great Lakes. This gives us three large dialect areas; Northern, Midland, and Southern, each with many subdialect areas. We arc most concerned with eastern New England and the rest of the North and the Tidewater and Piedmont areas in the South, areas which illustrate those dialect features that characterize various American dialect areas: 1. 2. 3. 4. 5.

The The The The The

loss of .'г/ before consonants and finally in words like car and barn. so-called broad-ah in words like half, path, and dance. short fa) of eastern New Englanders. ,''hw,''-/w/ variation between whales-Wales. various pronunciations of words like bad and cod.

Those features which characterize eastern New England and the Tidewater areas, (1)—(2), also characterize East Midland and Southeastern British English dialect areas which correspond to the source of the earliest settlers who coloni/cd New England and Virginia. The middle Atlantic states, on the other hand, which do not share characteristics with Eastern New England and the Tidewater, were settled by several groups, the most prestigious of whom, however, were probably the Quakers from northern England. If their social prestige influenced the speech of the area, I hen those Southern and East Midland English who did not have an /r/ would by imitation acquire one to imitate the /r/ dialect of Northerners. Northern British settlers in western New England and [he Great Lakes would have brought with them post-vocalic /r/'s; /a/'s instead of /э/*5 in words like hnt\ /a;/*s in /W/and dcmC€\ /hw/ instead o f / w / in white and whisper. These are the features which characterize the English spoken around the Great Lakes and in the Piedmont (though Piedmont speakers arc beginning to change

PHONOLOGICAL CHANGE! FROM OLD ENGLISH TO MODI-К N ENGLISH

359

because of the local social prestige of /r,'-less dialects). We could compare other features, such as the tempo of speech, oiher vowel sounds, and certain grammatical forms that would suggest further connections between British dialect areas and the original dialect areas in this country. In the Jast two centuries, American pronunciation has, of course, changed. New diphthongs have developed in the Southard North alike. But certain of our modern American patterns of pronunciation can be traced back to the dialect areas of seventeenth and eighteenth century England, dialect areas which themselves grew out of those in the Middle English period which, in turn, correspond to the OR dialect areas of Mercia, East and West Saxony, Kent and Norihumbria, areas that the Venerable Bede {673-735) thought reflected the continental origins of the Angles, Saxons and Jutes, whose dialects were allegedly scattered along the North Sea from the Rhine to Denmark—all of which is a continuity that testifies to the enduring nature of language as well as its constant tendency for change, (114)

Black

One final note about the history of English in this country concerns Black English. Jn dialect stories, racist jokes, old lime radio serials and movies, and so on. where dialect is supposed to be part of the humor or realism, we often read or hear Blacks allegedly speaking like this: mouf for mouth, bruvver for hen!her, dey for they, tick, for thick, tlebbit for devil, nuffin for nothing, fa* for far, ca' for cell, tmi for matt, *cep for except, jis Тот just, sicti for such, s'pose for suppose, fuss for first, and so on. Those individuals entirely ignorant of the history of language and how it changes have attributed these patterns to laziness, inherent stupidity, some even to the anatomical structure of Black speech organs. Others who believe themselves more charitable explain these patterns by claiming that many young Blacks never had the opportunity to learn "correct" pronunciation and that special programs of speech iherapy can eliminate the more obvious "faults," at least while these groups are speaking with speakers of standard dialects. It is worth noting first that virtually every one of these pronunciations characterizes various southern American dialeets and that each pronunciation can be found in various dialects of British English. Moreover, the kinds of assimilations and deletions that have produced these patterns have operated in Ihe language sinct Old English. This speech pattern illustrates two kinds of changes: substitution and loss. Initially, speakers of these dialects often substitute a voiced or voiceless slop for the corresponding spirant: they > e/ey, thing > ting, or jfj far №'•

Л60

(jHAMVfAR AND SOUND

Siting > frng. Medially, they substitute the corresponding labiodental spirant: nothing > nuffin, brother > hruvver. Finally, /f/ replaces /0/: mouth > motif, oath > oaf. Medially, jbj stereotypical ly replaces ,'v/: devil >
Because wich one exception, each change has sporadically occurred in many dialects of English, it is difficult to argue that any of these patterns are peculiar to Black English. The /d/ in Bedlam, for example, derives from the Щ in Bethlehem (the name of an insane asylum in London). More currently, the stop-for-spi rant characterizes the lower-middle class pronunciation of New York and Chicago, black and white alike: Dese tings is more often heard from older Chicago city officials than these things. English authors have frequently used these speech patterns to represent British class dialects: Charles Dickens wrote nujfin for natfang; William Thackeray wrote rw/j and motif for oat hi and month; In one of his plays, G. B. Shaw wroLe wiv for with. The substitution of/и/ for /rj/ in progressive participle endings, goto? fishin\ is attested to so often through the recent history of British and American Fnglish at all social levels that it is certain that at one time among some educated groups, it was the preferred pronunciation. Only the assiduous efforts of school leathers have changed the formal pronunciation of sittgBt' back to singing, an accomplishment of debatable value. The loss of the second of two consonants in words like test — tes7 cold — col are potentially more interesting because some have claimed that these are distinctively Black characteristics. Yet even the most superficial examination of seventeenth and eighteenth century spellings turns up literally hundreds of such spellings among educated writers. Other patterns reflect the speech of the Tidewater Engiish that has strongly influenced much Southern speech: The loss of pre-consonantal or final /r/: /ka/ for car, /ban/ for bam, and so on. But this is a British characteristic that was very likely brought here by the earliest settlers after the beginning of the loss of/r/. The only phonological pattern that can be plausibly attributed to the influence of a Black linguistic heritage might be the substitution of the stereotype /b/ for /v/ in words like debhit (devil) and bery (very). This change is not attested to in the phonological history of English in Great Britain, so it is not likely that it was brought here and passed on to slaves, [t is a change found in Neo-Melanesian and Chinese pidgins, but these are Pacific, not Atlantic pidgins. On the other hand, pidgins and Creoles based on Portuguese and French used in the Atlantic area do have jvj sounds, Voodoo, for example, is a Haitian Creole word (probably inherited from Ewe, a West African language, meaning demon or titulary god). So we cannot attribute the }vj > /b/ substitution to any of these Creoles. But when we push back beyond Creole and pidgin iangjages to the West African languages, we find that /v/ is not a common phoneme. Thojgh it is found in Ewe and Bini, it is not found in Yoruba, Наша, Efik, Twi, and for

PHONOLOGICAL CIIANUfcT FROM OLD ENGLISH TO MODERN ENGLISH

361

the most part, I bo. If we assume that the English pidgins learned by slaves reflected the phonological structure of their own native languages, much as a German speaker substitutes /v/ for /w/ because he has no native /w/ and substitutes the closest sound, so the majority of slaves perhaps had to substitute the sound they had closest to /v/. Most of their language did have a /b/ or /b/-like sound. The substitution of /t/ or /f/ for /6/, and the substitution of jdj for /S/ are frequently attested to in the history of British English, as noted above. But it is likely that the same change occurred in English pidgins and early Creoles, since neither /S/ nor /S/ are found in West African languages. Thus even though these changes can be found in British dialects, it is likely that they too can be attributed to the phonological structure of African languages interfering with the acquisition of English phonological patterns. (42, 54) Of course, none of this supports the claim that speakers of dialects other than what we have been calling standard English should be left uninformed of the differences between their pronunciations and the standard in their dialect area and of the social implications of using either pattern. Those speakers who as a rule drop final consonants in words like passed and jumped drop the past tense inflection; /рак/ and /jamp/ instead of /ptest/ and ftampt/. When this pronunciation is transferred to writing and a student writes He jump down from the roof instead of He jumped down from the roof., the consequences can be serious, because our written standard usage emphasizes very strongly not only correct spelling, but even пгшгс, correct verb forms. Unless a student can handle past tenses correctly (not to mention missing /l/'s as in He'll be ihere > Hebe there), he will be branded illiterate by educated middle class speakers and writers, " Illiterate' 4 in this sense, of course, means only "unable to use the dialect patterns of those who have the power to influence how others will spend their lives.*' The problem of education in this and future decades is to decide whether the social forces that stigmatize certain pronunciations are so strong that it would be irresponsible not to force students to learn the standard pronunciation of their dialect area. (This assumes that in fact we can give them a chance, whether they want one or not, to succeed later in a society that they might, at that moment, want nothing to do with.) The alternative, for the sake of supporting social identity and self-respect, is that the educational system should tacitly encourage the use of a speech pattern that later might contribute to social and occupational failure. Whatever reasons are used to support either decision, however, historical reasons are utterly irrelevant, either to support a pattern of speech that has its historical roots in earlier forms, or to reject a pattern that has never had the support of history. The decision is ultimately a political one that will have consequences on society, but more particularly, on the minds of millions of school children.

362

GRAMMAR AND SOL.'SD

SPELLING REFORM Considering the apparently confused relationships between sound and spelling it is not surprising that many have argued we should bring our spelling and pronunciation more closely together. English speaking children must spend large amounts of time learning to spell. Spelling has become a major social indicator of literacy (and hence potential social value), probably because it is so difficult and requires so much formal education. If we can devise rational alphabets for preliterale societies, we should be able to devise one for a society that his become more dependent on reading and writing than any other in mankind's history. Those with the least sense of how language works have argued that we should pronounce ;L word as we spell it. This usually takes (he form of coercing school children to "sound every vowel," of pronouncing a sentence like He always rejects her attempts to find the answer like (1) rather than like the more normal (2): 1, 2,

hi olwez rijekts har attempts tu faind 6i ffinsar / /hi Dlwiz rsjeks er atemps ta fain Йл a;nsor/

In other cases, it means pronouncing consonants that haven't been heard in standard spoken English for hundreds of years: The /t,' in Christmas or often, the fl) in balm, palm, and balk; the ;'d;' in graduate, the /z,1' in vision. Needless to say, this ltind of advice is worthless both because it is based on ignorance and because, fortunately, it usually has little or no effect on the way children speak once the final bell rings. With the exception of someone like Noah Webster, whose spelling book has sold millions of copies, no one person's advice about pronunciation makes much difference. The unfortunate consequences come when children grow up with a sense of guilt or inferiority over their * h bad" English, usually over a pronunciation that is in fact entirely correct. And worst of all, some of them remember well enough what such teachers had to say for them in turn to become linguistic martinets. Much more common but with equally little positive effect have been those who have tried to make spelling better reflect the way we speak. In recent years, spelling reformers have excited about the same degree of serious intellectual respect as vegetarians, Esperantists, and nudists. The beginnings of spelling reform go back to the Old English monks in their scriptoria silently changing an (a) to an <e> to reflect a vowel change. But al! these respcllings were unsystematic, idiosyncratic, and probably unconscious. The first change based on new principles of spelling was that of the late twelfthcentury monk, Orm. in his Ormulum, a versified version of the Gospels, he

PHONOLOGICAL CHANGE: FROM OLX) ENGLISH TO MOU£KIV ENGLISH

363

consistently doubled a Following consonant to indicate a preceding short vowel: annd all t=uss (this) ennglisshe hoc (book) iss orrmulum 3ehatenn (called) jet wile (will) ic shswenn (show) juw, forrwhi (why) l (gospel) is goddspcll nemmnedd (named). Unfortunately, this reform did not influence general scribal practice. Our habit of doubling consonants after a short vowel in two-syllable words: bitter as opposed to biter > is much mores inconsistent than his. The earliest Renaissance attempt at spelling reform was Sir John Cheke's translations of Matthew and Mark (c. 1550). It was в modest reform, m best, consisting only of doubling long vowels plus one or two other minor changes. The earliest serious treatise on spelling reform was that of Thomas Smith (1568), the first of several reformers who have taken the more radical step toward reform: He added new letters to represent consonants and accent marks to indicate long and shorL vowels. A year iater, 1569, John Hart followed with a new orthography of the same type. In 1580, William Buliokar and in I5S2, Richard Mulcaster proposed to reform spelling too, but with about the same effect as their predecessors. In the seventeenth century, a parade of names appeared to propose, new orthographies: Alexander Gill (1619), Charles Butler (1634), Simon Daines (1640), Richard Hodges (1643), JolTn Willdns Ц668). The eighteenth century was less actively reformist, partly because of the immense prestige of Samuel Johnson, who damped enthusiasm when he rhetorically wondered whether there was even arty point in trying to make an entire nation change its orthography if it would make its old boots useless and create extensive confusion among its current writers. But if Johnson's prestige may have effectively halted any serious attempts at reform in England until the middle of the next century, two Americans did propose vast reforms. As we might expect, Benjamin Franklin was one. In I7fi8t he published "A Scheme for a New Alphabet and Reformed Mode of Spelling." It consisted of a new phonetic alphabet that woutd fit pronunciation, and like n\l such attempts, it came to naught. Much more influential, indeed so influential Lhat many American dictionary makers invoke his name from the public domain, was Noah Webster (1758-1843). His fanciful theories on etymology and the genetic relationships of languages were without any foundation or subsequent influence. But his spelling books and dictionaries set new standards for American spelling and, some claim, pronunciation. Webster's early attempts at spelling reform were much more radical than Шояе that survive. After a very conservative approach to spelling in 1733, he recommended in 17&9 such spellings as bred, bitr> giv,

364

GRAMMAR AND SOUND

laf arkitecture, and obliik. He also recommended using various diacritics for both vowels and consonants. Most of these appeared in his first dictionary in \y.V\ A Compendious Dictionary о Г I\K English Language, But they began to disappear only a year later in his Л Dictionary of the English Language Compiled for the Use of Common Schools in the United State». By 1841, when he last revised his An American Dictionary of flic English Language, such spellings were very infrequent. It was his conservative approach to spelling reform in his immensely influential American Spelling Book that has resulted in many of the modest differences between American and British spelling, differences that for some reason buried in the recesses of their souls have driven some British pedagogues to flights of indignation usually reserved for those who abuse horses or slander the Queen. There are a few random differences: ntask-masquet chetb-cheque, jail-gaol, draft-draught, curb-kerb, plow-plough, story-storey, net-nett, gram-gramme, wagm-waggon, jewelery-jewellery, woolen-woollen, pajamaspyjamas, gray-grey, tire-tyre. The major systematic differences, most of which can be traced to Webster's influence, are these: 1. 2. 3. 4. 5. 6.

American (-or> vs. British <-our): hono[u)r, tobo{u)r, coh(it)r. American <-ll-> vs. British <-l->: travelled, Iesel(f)edt enrolled. American <-dg-) vs, British <-dge->: judgement, acknowledgement. American <-se) vs. British <-ce): defense-defence, pretense-pretence) license—licence, practice-practise (vb.). American <-ection) vs. British <-exion>: connection-comexion, infectioninflexion. American (-er) vs. British (-re): center-centre, meager-meagre. (225, 234)

If one has no better way to pass his time, he can debate the logic, efficiency, consistency, simplicity, or even the esthetics of these spellings. But finally, for either spelling, social contest will determine social consequences. Given two (admittedly artificial) sentences, we can predict how readers will respond, 1.

2,

The meagre grey colour of the gaol scenery at the centre of the theatre's stage is a reflexion tit'the bad judgement, of the licence given to fledgeling designers who may labour hard at their craft hut bring themselves little honour. The meager gray color of the jail scenery at the center of the theater's stage is a reflection of the bad judgment, of the license given to fledgling designers who may labor hard at their craft but bring themselves little honor.

PHONOLOGICAL CHANCE: т о м

OLD ENGLISH TO MODERN ENGI.ISH

365

It is highly probable that most educated Americans would consider the frit to be rather affected if written by another American, but if written by a Britisher, perfectly normal and therefore little if at all remarked. On the other hand, many, perhaps most educated Britishers would consider the second to be hopelessly illiterate if written by another Britisher and, if written by an American, proof of his irredeemably bumpkinesque background. While one might have personal reasons of orthographic integrity and so on for insisting on one spelling over another, the only valid reasons for preferring theater over theatre (or vice versa) arc social consequences. Beyond some spurious appeal to history, neither spelling has any sanction except custom. After a century lag, spelling reform became a popular issue again in Britain when in the middle of the nineteenth century, Isaac Pitman devised a shorthand system based on phonetic principles. Since that time, the reformer ranks have swelled with the names of such respected scholars and writers as Walter Skeat, Gilbert Murray, Daniel Jones, Harold Orion, Robert Bridges, and most prominently, perhaps, George Bernard Shaw, who left a substantial endowment to evangelize spelling reform. In the United States, W, D. Whitney, one of the foremost American philologists of the nineteenth century, encouraged spelling reform. In 1875, the American Philological Association created a committee to investigate the project. From 1906 to 1919, Andrew Carnegie financially supported the Simplified Spelling Board- President Theodore Roosevelt supported its efforts to the extent of ordering public documents to use new spellings in certain words. Congress reacted predictably and there has been no official interest in spelling reform since. In the early thirties the Chicago Tribune began to champion reformed spelling in earnest, announcing several new spellings like agasl for aghast, Umd for island, tho for though, crutn for crumb, Uiher for leather. By 1973, the list had shrunk to tho and thru (and their combinations) and some others as innocuous as catalog for catalogue and cigaret for cigarette. In recent years, most of the scholarly effort has been devoted to creating alphabets for languages that do not have writing systems and to the teaching of reading to young children. The latter endeavor is based on the assumption that if children can learn to read quickly and easily with a simplified alphabet, (hey can then be gradually weaned to conventional orthography. In this way, they need not endure the Oh, Ok, See, See stage any longer than necessary. Since all experiments of this kind are judged successes by those who devise them, we will have to wait until the method is used by many teachers with no vested interest in the project before we can call it a success. The objections lo spelling reform arc legion, though most of them are trivial. The desire to retain the etymologies signalled through traditional spelling hardly equals the pain children endure when they try to learn to read. The claim that phonetic spelling would eliminate one way to distinguish homophones is without merit since it is virtually impossible to think of any

366

URAMMAR AND SOUND

likely stretch of continuous discourse in which we could confuse knot and not, bow and bttugh, sea and see, night and knight, son and sun., hare and bear,

two, too, or to, and so on. The objection that spellings like cewft nife, skweez, and £>Лл/, are unfamiliar, distracting or ugly is beyond rational discourse. Many of us are unable to repress a shudder at signs advertising Lhe Kreemi-krunch Kawfi Kav/ttier. But we must ask whether our eyes are so sensitive that shielding them from such orthographic horrors justifies the weekly spelling test or the occupational problems that dog the poor speller. In the eighteenth century, after all, critic probably offended the eyes of as many Englishmen as hotter and center do today. (63, 92, 178, 182, 225, 245) There are some objections to spelling reform that are more persuasive, however, particularly in a polydialectal society like our own. For example, a sentence like I found a car it) the creek could be represented as either: 1. 2.

/a feun э ka in бэ krik/ /ay faund э каг in оэ krik/

We could simply say that the dialeet with the greatest number of speakers would be the one on which to base a new spelling. But it would still be a minority dialed since no one dialect is uniformly spoken by a majority of speakers in this country. Indeed, the fact that our spelling is so loosely tied to pronunciation allows speakers from all English dialect areas from England to the United States to Australia to use the same symbol for somewhat different pronunciations. This does not argue for alternative spellings of /ad/ such as cue, sight, buy, by, lie, neither, and so on. But it does point up the problems in trying to select one unambiguous symbol for one sound, and the value of having one symbol for different dialectal pronunciations of the same word. The alternative to excluding large numbers of speakers from the spelling system would be to have multiple spelling systems for different areas, a condition infinitely less desirable than an inefficient spelling system. One other kind of systematic relationship between our spelling and pronunciation should also be noted, both because it reflects on the history of our language and because it is another case that testifies to how efficient a nonphonetic spelling can be. As we have abundantly seen, English spelling is not perfectly phonemic. But we could argue that given the variable pronunciation of many Romanee words, the system should under any circumstances not be phonemic. If it were, tiien the vowels in boldface in these pairs of words would have to be spcitcd differently; abdomen-abdominal, aceede-accssston, adjectice-adjeclkal, angel-angelic, biie-biiious, ctiasie-cliasiity, сапе-conic, «lostic-tUtsActty, infunt-infanti-

PHONOLOGICAL CHANGE: FHOM OLD ENOLISH TO MODERN ENGUSH

367

t, chit-civility, pu/iteh-panitive, sign-significant, title-titular. How we spell these sets of words relates to how we pronounce them in thiee ways. (1) Under light stress, all vowels become a kind of obscured midcentral vowel, /э/, or a high-central vowel, symbolized (if, the vowel of unstiessedjust in just П minute or in the inflection of Jilted. Thus for this vowel, spelling is irrelevant, once we learn various kinds of stress rules that reduce any full vowel to an obscured vowel. If the two words, cone and conkity, were spelled kon and ksnisity, then given only ksniMty, we would not know from spelling alone the vowel of Lhc root word. It could be an as in the first unstressed vowel in civility, or an < 15) as in the first unstressed vowel in remedial, or it could be an as in the unstressed vowel in alacrity. (2) The second kind of relationship pairs stressed and lax vowels. If we rnaLch the stressed and lax vowels /6/-/3/ and /п/-/э/, then certain sets of words regularly vary between them, depending on the suffix: assumeassumpt ion, verbose-verbosity. (3) A second kind of tense-lax relationship partially mirrors the Great Vowel Shift:

•

/ai/-/i/: /I/-/e/: /ё/-/зе/: /au/^/э/: /5/—/a/;

bile-bilious acceed-accession chaste-chastity profound-profundity cone-conic

In some cases, one (or two) letters represent three different segments: /a i/-/i/-/a/; /i/-/e/-/a/: /с |'-/зе/-/э/: /5/-/а/-/э/: /й/-/а/-/э/:

anal_y zc-ап al^tical-analysis repftjt-repetitive-reptftition expldi'n-expltf natory-expl a nation cone-can ic-c
In each of these three sets of words, we would need different letters if we followed a strictly phonemic spelling system. What English has in sueh wt?rds is not a phonemic spelling system but a morpho&rapheniic or rjiorjihophoiieinic spelling system. That is, given certain stress patterns and classes of suffixes, the letter or letters represent an underlying form that is realized in varying but predictable ways. Give» the nonceconelete, we know we can generate condeiition /kanklstiSan/ and /kankletstiv/, just as we can, given, Everyone knew that he left, generate that he left was known by everyone. The letter actually represents an

368

GRAMMAR AND SOUND

abstract phonetic deep structure that can be manifested in various kinds of surface structures. The rules which we have to learn in order to manifest these deep structures as sound have been proposed at length by generative phonologists. But since the most profound generalivc analysis of English sound patterns is over 400 pages long, it is impossible to do justice here to the powerful insights these linguists have had into English phonology. (33) The same kinds of underlying morphogtaphemic or morphophonemic entities are relevant to consonant changes. Compare these: /k/-/s/: electric-electricity /k/-,'§/: logic-logician jtj-lif; divert-dhersioti /t/-/S/: insert-insertion /z/-/s/: analyze-analysis /t/-/E/: right-righteous tf/-/g/: alkge-nllegution jsj-jsj: impress-impression In the same way, the first word in pairs like bomb-bombard, resign-resigntttion, damn-damnation spell an underlying segment, in bomb, in resign, and
REFERENCES

*1.

Alltn, Virginia H, "Teaching Standard English as a Second Language," The Teachers College Record, 68:355-70 (1967). 2. Anderson, George K. The Literature of the Anglo-Saxons. Princeton, N J., 1949. 3. Anttila, Kaimo. An introduction to Historical and Comparative Linguistics. Mew York, 1972. 4. AiYtaod, E. Bagby. A Survey of Verb Farms in the Eastern United States, Ann Arbor, Mich., 1953. 5- Bam has, Rudolph С " Verb Forma in -s and -lh in Early Modem English," Journal of English and Germanic Phlhlogy, 4б:Ш- 87 (1947). 6. barber, Charles. Linguistic Change in Present-Day English. London, 1966, 7. Иа ugh, Albert С A History of the English Language. 2nd ed. N ew Y ork, 1957. 8. Bender, Harold H, The Home af the indo-Europeans. Princeton, N.J,, 1922. 9. Berlin, Brenl and Paul Kay, Basic Color Terms, Their Universality and Evolution. Los Angeles and Berkeley, 1969. Ю- Berndi, Rolf. ' T h e Linguistic Situation in England from the Norman Conquest to the Loss of Normandy (1066-1204),*' Phitvtogica Pragensia, 8:145-63 (1965). П. Bjorkman, E, Sr.utiilitiavkin Lean- Words in Middle English. Malic, 1900-2. 12. Blair, Peter H, An Introduction to Anglo-Saxon England- Cambridge 1?56369

370 13,

. Roman Britain and Early England, 55 B.C.-A.D, 1963. 14, liJrmrnfii-hl, Leonard. Language. New York, 1933.

971, Edinburgh,

15, liknimlirlLl. M<jrii>ii W. and Leonard Newmurk. A Linguistic Introduction tv the History of English, New- York, 1963. J6. Bulinger. Dwiuht. "Rime, Assonance, and Morpheme Analysis," Word, 6:117-36(1950). П, Иг Он I, Michel. Semantics, Studies in the Science of Meaning (trans. Nina Cust). New York, 1964, (originally pub. 1900). IS. Brook, G. L. English Dialects. London, 1963. 19. Brown, Roger, Abraham Black, and Arnold Huruwiiz. "Phonetic Symbolism in Natural Languages,"Jtwnmt of Abnormal and Social Psychology, 50:38893 (.1955). 20. and Marguerite Ford. "Address in American English," Journal of Abnormal and Social Psychology, 62: 375-85 (1961), 21.

— and Albert Gil man. "Pronouns of Power and Solidarity," in Styte in Language. Thomas A. Sebenk, ed. Cambridge, Mass., 1966.

22. Bninncr, KarJ. "Hxpanded Verbal ? : onvisin tarly Modern tnglish," English Studies, 36:215-21 (1955). 23. Duck, Carl D. A Dictionary of Selected Synonyms in the Principal IndoEuropean Language*. Chicago, 1949. 24. Byrne, Slsrer S<, GwaMh». Shakespeare's U*e of the Pronoun of Address, Its Significance in Characterization and Motivation. Washington. D.C., Catholic University of America. I'hD Dissertation 36, 1936, 25. Campbell, Л, Old English Grammar. Oxford, 1959, 26. Carr, С. Т. Nominal Compounds in Germanic. London, 1939. 27. Carr, E. H. What Is History? New York, 1967. 28. Cawidy, Frederic G. and Richard N. Ringer, eds, Brigftt's Old English Grammar and Header. 3rd ed. New York, 1971. 29. СпяГе, Wallace, Meaning and the Structure ofLunguage, Chicago, 1970. 3d, Chambers, R. W, and Marjnrle ij-nmi. A Book of London English, 1384-1425. London, 1931. 3J. Chomsky, Noam. Aspects of the Theory of Syntax. Cambridge, Mass., 1965. 32. 33.

. Syntactic Structures. The Hague, 1957. and Morris Halle. The Sound Pattern of English. New York, 1968.

34. Culltngwiwd, R, G. Roman Britain. New York, 1932. 35. Curme, George O. English Grammar. New York, 1966, 36.

, " A History of English Relative Constrjctions," Journal of English and Germanic Philology; 11:10 29, 180-204, 355-80 (19! 2).

37. liit ;imp, David. " T h e Genesis of the Old English Dialects: A New Hypothesis," Language, 34:232-44 (195Й). 38. Dcese, James, the Baltimore, 1965,

Structure of Associations in Language and Thought.

REFERENCES

371

39. Dillard, J. L. Rfock English. New York, 1972. 40. • -. " O n the Beginning* Ы Black English in ihc New Wurld," Orbis 21:523-36(1972). 41. Dlrineer, П, The Alphabet. 2vu]s. New Vork, 1968. 42. Dobson, Fl, J , "Harly Modern Standard Eoglhb," Transactions of the Philological Society, 1955:25-54, 43-

. English Pronunciation, 15(M)-1700. 2 vols. Oxford, 1957,

44. Dyen, Isidore. "Lcxicostatistics in Comparative Linguistics,'* Lingua, 13: 23U-39 (1965). 45. Оукслщ, Karl W, "Where Our Grammar Came From," College English, 22:455-65(1961). 46. Kkwall, Eilert. Studies on the Genitive of Groups in English. Lund, 1943, 47. Ellegfat, Alvar. ГА? Auxiliary " Do," The Establishment and Regulation of Its Use in English, Stockholm, i953. 48. Kmerson, O. F. "English or French in ihc Time of hdward III," Romanic Review, 7:127-43 (1916), 49. EBgblom, Victor. " O n the Origin and Early Development of the А и xiliary ' D o ' , " Lund Studies in English, б (1MB), 50. ENti idi. Robert M. und Hans Sperber, Three Keys to Language. New York,

1552. 51. Evans, W. W. " T h e Dramatic Use of the Second Регяоп Singular Pronoun in Sir Go wain and the Green Knight," Stttdia Neophifofoglca, 39:38—45 (1967). 52. Farmer, John S. and William Krnesl Ili-nki. Shtng and Its Analogues. 7 vols, London. 1890 1904. (Republished in one volume, New York, 1970,) 53. Id Mild, Ralph, "Tense and the Form BE in Black English," Language, 45: 763-76 (1969). 54. • — and Walter Wolfram. "Some Lingjktic Features of Negro Dialect," in Teai-hing Standard English in the Inner City. Ralph W. FasoJd and Roger W. Shuy, cds. Washington, DC, 1970. 55. h'ilhnore. Charles J. " T h e Case for Case, ! ' in Universals in Linguistic Theory, Emmon Bach and Rohert T, llai'ms. eds. New York, 1968. 56. Fischer, John 1,, "Social Influences in the Choice оГа Linguistic Varianl," Word. 14:47-56(1958), 57. Ferster, M a s . Keltisches Wvrtgut im Englischen. Halle, 1921. 58. FoK
372

REFERENCES

64. Frits, Charles C, "Lingullies. The Study of Language," Chapter 2 in Linguistics and Reading, New York, 1964. 65.

. " O r the Development of the Structural Use of Word-Order in Modern English," Language, 16:199 20Й (1940).

66.

, The Structure of English, An Introduction to the Construction of English Sentences. New York, 1952.

67. Gaaf, W. van der. The Transition from (he Impersonal to the Personal Construction in Middle English. Heidelberg, 1904. 68. Gardner, R, Alien and Beatrice T. Gardner, "Teaching Sign Language to a Chimpanzee," Science. 165:664-72 (!%9), 69. Gelb, I, J. A Study of Writing. 2nd ed. Chicago, 1969. 70- Gleason, I knry A. introduction to Descriptive Linguistics, revised ed., New York, 1961. 71. Gondall, Jane. "Chimpanzees of ihe tiombe Stream Reserve," in Primate Behavior, Field Studies of Monkeys and Apes, I, DeVore, ed. New York, 1965. 72. Gnodenough, Ward H. "Componential Analysis and the Study of Meaning," Language, 32:195 -216 (1956). 73. Greenberg, Joseph H., cd. Universal* of Language. 2nd ed, Cambridge, Mass., 1966. 74. Grettinujjh, Jamts B. and Gcorqe I.. Kiiircd^c. Wards and Their Ways in English Speech. New York, 1901. 75. Grose, Francis. /811 Dictionary of the Vulgar Tongue. London, 1811. (Reprinted Chicago, 1971). 76. Hall, John R. C\ A Concise Anglo-Saxon Dictionary, with й Supplement by Herbert D. Meriit. 4th cd. Cam bridge, 1960. 77. Hall, Robert Л. Pidgin and Creole Languages. Jthaca, N.Y., 1966. 78. Hams, Zellig S. Methods in Structural Linguistics. Chicago, 195!. 79. l lurmii-n, Charles v. " T h e Persistence of Tradition in (irammar," Quarterly Journal of Speech, 48:174-86 (1962). fiO. Haugen, Einar. " T h e Analysis of Linguistic Borrowing," Language, 26: 21031 (1950). 81. Hefftier, Rtic-MtrilL General Phonetics. Madison, Wise., 1949. 82. Hewes, Gordon W. "Primate Communication and the Gestural Origin of Laneuage," Current Anthropology^ 14:5 24 (I97J). 83.

. Language Origins, a Bibliography. Boulder, Colo., 1971.

Й4. HilJ, Archibald A. Introduction to Linguistic Structures. New York. !95885.

. "Phonetic and Phonemic Change," Language, 12:15-22 (1936).

86. llockett, Charles b\ A Course in Modern Linguistic*. New York, 1958. 87.

. " The Origin of Speech," Scientific American, September, 1960:3-11 -

88.

. " T h t Problem of Universal in Language," in Oreenberg, 1-29.

89.

. " T h e Stressed Syllables of Old English," Lattguegt, 35:575-97 (1959).

BQPF.RENCES

90.

373

and Robert Aschcr. " The Human Revolution," Current Anthropology 5:135 6S(!964

91, Hoerfgvsnld, Henry M, Language Change and Linguistic Reconstruction. Chicago, i960. 92. Hulbert, James R. DtettoMlies, British and American, London, 1955, Jackson, Kennel It. Language find F Us/orу in Early Britain. Edinburgh, 1953. 94. Jacobs, Ruderick Л. and Peter S. Roscnhaum. English Transformational Grammar. WalLham. Mass., 3 973. 95. Jcsperwa, Ottu, Growth and Structure of the English Language Oxford, 1954 96. 97.

-

9th ed

'.Language, Its Nature, Development mid Origin. New York, 1922,

— . A Modern English Grammar on Historical Principles. 7 vols, Copenhagen, 1909-^9.

96- Jones, Charles. An Introduction to Middle English. New York, 1972. 99. Innes, Richard. The Triumph of the English Language. Stanford, Cal., 1953. 100. JouS, Martin, " T h e Five Clocks," International Journal of American Linguistics, 28:2(1962). 101, Katz, J . J. Semantic Theory. New York, 1972. 102. and J . A. Fodor, " T h e Structure of a Semantic Theory," Language, 39:170 21 103. Kendriclis, T. D. A History of the Vikings. New York, 1930. 104. Kennedy, Arthur G. A Bibliography of Writing* on the Eltgtlsh Language from the &i'x inning of Flint ing to the End of 1922, Cam bridge, Mass, and New Haven, Conn., 1927. 105, . The Pronouns of Address in English Literature of the Thirteenth Century. Stanford, Cal., 1915. 106. KinR, Robert D, iiiaioricu! Linguistic* and Gen&ettee Gmnmiar. tnglewood Cliffs, N.J., 1969. 107. Ktifceritz, Helge. " Elizabethan Prosody and Historical Phonology," Annales Acadamiue Regitif> Scienticirmn UpxuHcnst.t, 5:79-102 (1961). 108, , A Guide to Chaucer's Pronunciation, New York, 1962. 109.

-. Shakespeare's Pronunciation. New Haven, 1 Ю. Krapp, Genrgc P, The English Langnctgr in America. 2 vols. New York, 1925. 111. Kuhn, Sherman M. " O n the Syllabic Phonemes of Old English,'4 Language, 37:522-38(196!).

И 2. Kuhn, Thumas. Tin.' Structure of Scientific Revolutions. 2nd ed. international Encyclopedia of Unified Science. Vol. 2-2. Chicago, 1*70. ИЗ. Kurath, Hans. 'The Loss of Long Consonants and the R-isc of Voiced Fricatives in Middle English," language, 32:435^15 (1956). — — , " T h e Origin of the Dialectal Differences in Spoken American English/' Modern Philology, 25:3*5-95 (1923).

П5, — — ,

Ll phonemic* and Phonics in Historical Phonology," American Speech, 36: У3-100(1961).

374

RITFERKNCES

116.--

- and Sherman M. Kuhn. cds. Middle Englhh Dictionary. Ann Arbor,

Mich,, 1954— H7 . | Raven I. Me Dai id, Jr. The Pronunciation of English in the Afternic а ш States. Ann Arbor, Miclv, 1961. П8, Labov, William. " T h e Social Motivation of a Sound Change," Ward, 19; П9, — — . "Contraction, Deletion, and Inherent Variability оГ the English Copula," Language, 45:715-62 (1969) 120, •. " O n ihe Mechanism of Linguistic Change." Monograph Series on Languages and Linguistics, 18:91-114 (1965). 121. . The Social Stratification of English in New York City. Washington, D.C., 1966, "The Study of Language in Its Social Setting." Stodium Generate, 23:30-87(1970). 123. Lancaster, Jane. "Primate Communication Systems and the Emergence of Human Language," in Primates, Studies in Adaptation and Variation, Phyllis Jay, ed. New York. 1968. 122.

124. Less, Rngtr, ed. Approaches to English Historical Linguistics. New York. 1969. 125. Let. Donald W. Functional Change in Early English. Menasha, Wise,, 1948, 126. Lees, Robert B. The Grammar of English Nominulizations. Internal ion ai Journal of American Linguistics, Publication 12, Bloomington, lnd., 19<5O. 127. Legge, M. Dominica. ''Anglo-Norman and the Historian," History, N.S. 26:163-75(1941). 128. LcKniann, Winfrcd P. Historical Linguistics, an introduction. 2nd ed,, New York, 1973. 129. and Vakov Malkivl, ed*. Directions for historical Linguistics. Austin, Texas, I96S. 130. Lennehcr", Eric. Biological Foundations of Language. New York, 1967. 131. Leonard, Sterling, The Doctrine of Correctness in English Uwgf. 17QQ-I80Q. Madison, Wise, 1929, 132. Lester, Mart, introductory Transformational Grammar of English. New York. 1971. 133. Loflin, Marvin D, "Negro Nonstandard and Standard English: Same or DilTerent Deep Structure?" Orbis, 18:74-91 U969). 134. Long, Ralph B, The Sentence and Its Paris. Chicago, 1961, 135. Mackenzie, B. A. The Early London Dialect. Oxford, 1928. 136. Mart-hand, Hans. The Categories and Types of Present-Day Engfith Word Formation, и Synchronic-Diachrank Approach. Wiesbaden, I960. 137. Marckwardt, Alhcrt H. American English, New York, 1958. ПЙ. Mathews, \lilfwd M. The Beginnings of American English. Chicago, 1931. 139.

, ed. A Dictionary of Americanisms on Historical Principles. 2 vols. Chicago, 1951 (republished in one voL, 1956).

REFERENCES

375

140. Matlhews, William. Cockney Past and Present, a Short History of the Dialect qf London. New York, T938. 14)- McLtavid, Haven 1. Jr. "Postvocalic f-rt in South Carolina: A Social Analysis," American Speech. 23:194-203 (1948). 142. Mcknight, George H. English Words and Their Background. New York, 1923. 143. McLaiiKhlJii, John C. Aspects of the History of English. Ne^v York, 1970. 144. Meillet, Antoine. The Indo-European Dialects (trass. Samuel N. Rosenberg). Alabaina Linguistic and Philological Series 15. University, Ala,, 1967. J45. Mencken, H, L. The American Language. 4th ed. and two supplements, abridged with annotations and new material by Raven I. MoDavid, Jr., with the assistance of David W. Maurcr. New York, 1963. 146. Moore, Samuel. "Earliest Morphological Changes in Middle English," Language, 4:2Э8-Й6 (192?). 147.

. " Loss of Final n in Inflectional Syllables of Middle English," Language, 3:232-59 (1927).

148.

Historical Outlines of English Sounds and Inflections, revised by Albert H, Marckwardt. Ann A r l w , Mich,, 1951.

149.

, Sanford B, Mcech, and UarotU Whh*haJI. 'Middle English Dialect Characteristics and Diaket Boundaries." Essays and Studies in English and Comparative Literature. University of Michigan Publications, Language and Literature. 13:1-60(1935).

150. Mosse, Femand, A Handbook of Middle English (trans. James A. Walter). Baltimore, \952. 15^. . Hi-Wiire de In forme piriphraitiique it re + participe en present en Gerniaiiit/ue. 2 parts. Paris, 19Ж 152. Moulton, William G. l 'Thc Stops and Spirants of Early tiermanic," Lot* 153. Murray, James, A. J),, Sir. Evotutivn of English Lexicography. Oxford, 1900. 154. Muaianoia, Tauno F. A Middle English Synwx, Part 1; Parts of Speech. Helsinki, 1940. 155. Norman, Nathan, " Pronouns of Address in the Canterbury Tales," Mediaeval Studies, 21:193-201 (1359), 156. Ogden, C. K. and I. A. Richards. The Meaning of Meaning. New York, 1923. 157. Ossood, С. Б., G. J . Suci, and P. H. Tannenbaum. The Measurement of Meaning. Urbana, 11L, 1957. 158. The Oxford English Dictionary. A. H. Murray, H. Bradky, W. A, Craigie, and С. Т, Onions, eds, 13 vols. Oxford 1933. Originally published 1Й84-192Й as A AW English Dictionary on Historical Principles, reissued with Supplement in 1933. Supplementval !, A O. R. W. Buiehficld, ed. Oxford, 1972. 159. Partridge, KrEc, ed. A Dictionary of Slang and Unconventional English. 5th ed. Mew York, J967. 160. . Name into Word. New York, 1950, 161. , ed. Origins, A Short Etymological Dictionary of Modern English. 2nd cd. New York, 1959.

376

REFERENCES

162. . Slang Today and Yesterday. 3rd ed. New York, 1950. 163. I'edersen, 11. The Discovery oj'Language (Iraus, John W. Spargo). Bloonnington, Ind., I9fi2. (Originally published as Linguistic Science in the '9th Century, Methods and Results. Cam bridge, Mass., 1931.) 164. Fenzl, Herbert. " T h e Evidence for Phonemic Changes," in Studies Presented tv Joshua Wkaimough, Ernst Pulgram, ed. The Hague, 1957. 165. Pound, Louise. Bicmis, Their Relation to English Word Formation. Heidelberg, 1914, 166. Poutsma, О. Л Grammar of Late Modern English. 4 vols. Gronineen, 19041926. {Part I, 2nd ed., in 2 sections, 1928, 1929). 167. Praz, Mario. " T h e Kalian Element in English," Essays and Studies,) 5:20-66 (1929). 169, Preinack, David. ''Language in Chimpanzee?" Science, П2:808-22 (1971). 169, Price, H. T- Foreign Influences on Middle English. Ann Arbor, Mich., 1947. 170,

. A History of Ablaut in Jhe Strong Verbl from Caxton to the End of the Elizabethan Period, Bonn, 1910.

171. Puisram, Ernst. "Family Tree, Wave Theory and Dialectology," Qrbis, 2: 67-72 (1953). 172. Pylts, ITwmas. The Origins and Development of the English Language. 2nd ed. New York, 1971, 173. -. Wards and Wayt of American English. New York, 1952, 174. Quirk, Randolph anil Sidney Otcnhaum. A Concise Grammar of Contemporary English, New York, 1973. 175. — — and С L. Wrenn, An Old English Grammar. Ntw York, 1957, 176. The Random House Dictionary of the American Language. New York, 1966, 177. Ranlavaam, Irma. " O n the Development of the Periphrastic Dative in Late Middle English Prose,' 1 Meuepnitalogischc Mitttttungen, 63:175-203 (1962). 178. Read, \llon Walker. "American Projects for an Academy to Regulate Speech," Publication of the Modem Language Association, 51:114J—79 (1936), 179. . " T h e Speech of Negroes in Colonial America," The Journal of Negro History, 24:247-58 (1939), 180. Reed. Carroll E. Dialects of American English. Cleveland, 1967. Ш . Roberts, A. Hood. A Statistical Linguistic Analysis of America» English. The Hague, 1965. 182. Robertson, Stuart. The Development of Modern English. 2nd ed. Revised by Frederic G. Cassidy, EngLewood Cliffs, N.J., 1954. 183. Robins, R. H. A Short History of Linguistics. Bioomington, 1968. 184. Rorty, Richard. The Linguistic. Tarn. Chicago, 1967, 1Я5 Ross, Alan S. C. " T h e Origin of (he «-endings of the Present Indicative in English," Journal of English ami Germanic Philology, 33:68-73 (L934). 186, Runibaugh, Duane M-, Timothy V. Gill, and E. С von GrlaMffiM. " Heading and Sentence Completion by a Chimpanzee ( Р а т ) Г Sciencet 182:731-33 (1973),

REFER ENCRS

377

187. Russell, J«tnkh C. British Medieval Population, Albuquerque, 1948, 188. Samuels, M. L. Linguistic Evolution wish Special Reference to English, Cambridge, 1972. 189. $eluf«uder, H, Pejorative Sense Development in English. Groningen, 1929, 190. Scolt, C h a r t s T, and Jun L. Eriekson, cds. Readings for the History of the English Language, Boston, I96S, 191. Searlt, John, ed. The Philosophy of language. Oxford, 1971, J92. Sebeolf, Thomas A,, ed- Current Trends in Linguistics. Vol. III. The Hague

IMS,

193. •

— , ed. Style in Language. Cambridge, Mass., 1%Ь,

194. Serjeantson, Mary S. A History of Foreign Words in English. London, 1935, 195. Shannon, Ann. A Detcrlptlve Syntax of the Parker Manuscript of the AngloSaxon Chronicle from 734 to 897. The Hague. 1964. 196. Shelley, Percy. English and French in England, 1066-1100- Philadelphia, 1921. 197. Shuy, Rogti\ Discovering American Dialects. Champaign, III, 1967. 19S, Skeat, Walter W, English Dialects from the Eighth Century to the Present Day. Cambridge, 1911. 199. Skinner, B. F. Verhal Behavior, New York, 1957. 200. Sleild, James. "Bi-tlmlcctalism: The Linguistics of While Supremacy," English Journal, 58:1307-29 (I9fi9), 201. — and Wlma R. Ebbit, eds. Dictionaries and THAT Chicago, 1962. 2(J2.

Dictionary.

and Gwin J. Kolb. Dr. Johnson's Dictionary; Essays in the Biography of a Book. Chicago, 1955.

203. Slobin, Dan. Psycholinguistics. New York, 1971. 204. Sumlmnrth, Kranklin С "Fondly Trtc Diagrams," Language, 40:557-65 (1964). 205. S t a r n « , D. '!'. and (Jcrlmdc IVoycs. The Englhh Dictionary from Cawdrey to Johnson, 1604-1755. Chapel Hill, N.C. t 1946. 206. Stenton, F. M. Anglo-Saxon England, 2nd ed. Oxford, 1947. 207. SleinbHE, Danny D. and Leon Jakobovitz. Semantics, an Interdisciplinary Header in Philosophy, Linguistics, and Psychology. Cambridge, 1971. 20S. Stern, Gustaf. Meaning and Change of Meaning with Special Reference to tiit- English Language. Goteborg, Sweden, 1931, 209. Sterlet, Robert D. English and Its History, the Evolution of a Language, Boston, 196S. 210.

. " T h e Morphemic Evolution of Middle English She," Studies, 45:381-Й8(1У64).

English

211. Stewart, William. "Continuity and Change in American Negro Dialects," Florida Foreign Language Reporter, 6,2:3-4, 14-16, 18 0968). 212.

•—-, "SociolinsuLsiic Factors in the History nf American Ne£ r c Dialects." Florida Foreign Language Reporter, 5,2:11, 22, 24, 26, 30 (1967).

378 213. Stockwell, Robert P. "The Middle English 'Long Close' and 'Long Open' Mid Vowels." Texas

214. 215.

Studies

in Literature

and Language,

2:259-6S (1961).

, "Mirrors in the History of English Pronunciation," in Lass. and Ronald S. K. Macaulay, eels. Linguistic Change and Generative Theory,

B l o n m i n g t o n . Jnd., 197].

216. Suggetl, Fltlen. "The Use of French in England in the Later Middle Ages," Transactions of the Royal Historical Society, 4th scries, 28:61-83 (1946), 217. Swieczkowski, Wnicrian. Word Order Patterning in Middle English, a Quantitative: Study Based on Piers Plowman and Middle English Sermons. The Hague, 1961. 218. Thieme, Pawl- "The Indo-European Language," Scientific American, October, i 955:63-74. 219. Ttagcr, George L, and Henry Let Smith, Jr. An Outline of English Structure. Studies in Linguistics, Occasional Papers 3. Noiman, Okla., J95• 220. Ttaugott, Elizabeth Closs. " Diachronic Syntax and Generative Grammar," Language, 41:402-15 (J965). 221. . A History of English Syntax. New York, 1972. 222. , "Toward a Theory of Syntactic Change," Lingua, 23:1-27 (1969V 223. Trcveiyan. G. M, History of England, Volume Orm: From the Earliest Times to the Reformation. New York, 1966224. Turner, Lorenzo D. Africanisms in the Gullak Dialect. Chicago, ]949. 225. Vallins, С, Н. Spelling. London, 1954. 226. . The Pattern of English. London, J956. 227. Van Dongen, G. A, Ameliorativcs in English. Roller daw. 1933228. Venciky, Richard L. The Structure of English Orthography. The Hague, 1970, 229. Vising, Johan. Anglo-Norman i.angnage and Literature. London, J923. 230. Visser, F- Th. An Historical Syntax of the English Language. 3 vols. Leiden, 1963-69. 231. Wagner, Karl. Generative Grammar Studies in the Old English Language. Heidelberg, 1969. 232. Walcuti, С С. ( i Thc Pronouns of Address in Troilux and Criseyde," Philological Quarterly, 14:282-87 (1935>. 233. Waldron, R. A, Sense and Sense Deoelopment. New York, 1967. 234. Warfel, Harry R, Noah Webster, Schoolmaster to America. New York, 1936. 235. Watsnn, J. B. ''Psychology as the Behaviorisu View It," Psychological Review, 20:158-77 (1913). 236. Webster's Third New International Dictionary, Springfield, Mass,. J96I237. WeLnreieh, Uriel, " Explorations in Semantic Theory, in Current Trends in Linguistics, Vol. III. Thomas A. Sebeok, ed. The Hague, 1966. 238. . Languages in Contact, Findings and Problems. The Hague, 1968. 239. , WilJiani Laboy and M, I, Кепок- " Empirical Foundations for a Theory of Language Change," in Lebmana and Malkicl.

BEFERENCES 379

240. WeatiWKth, Harold and Stuart Berg Fie* пег, eds. Dictionary of American Slang. N&w York, J960. 24J- Wilson, R, M. "English and French in England, 1100-1300," History, N.S. 28:37-60(1943), 242. Woodbine, George. l The Language of English Law," Speculum, If- 395-436 (1943). 243. Wright, Joseph, ed. The English Dialed Dictionary. 6 vols. London, ]39&1905. 244. Wyld, H, С. Л History of Modern Colloquial English. 3rd ed. Oxford, 19Э6. 245- Zachrissun, R. E. " F o j r Hundred Years of English Spelling Reform," Stadia Neophilalogicu, 4:1-69 (193J).

INDEX OF MODERN ENGLISH WORDS

a, 112, 163, 226,239242, 290 aardvaj'fc, 101 abbey, 81 abbot, $1 abdomen, 200, 366 abdomina!, 366 atbduct, 114 abet, 140 abeyance, 133 abide, 260, 261 abject, B7 able. В2 aboard, 265 abominable, 207 abominably, !91 abstract, 1S4 academic. 116 acctde, 366, lb~ accent, 317 accept, 32: acte^ion, 366, 367 accident, 172, 255 actloy. 1-40 accordant. 88 aazosl, 172 account, 114, 187 accrue. 135 accuracy, 129 accuse, 157 acid, 209, 210 acknowledge. 114, 157 j

acknowledgment, 364 acquit, 136 acquittance, 135 acre, 320 acrid, 209.210 atrouym, 204 acrftsb, 28 acted, 115 acting. L25 n, K2, 125, 157

activate, 129 sets. 125 actual, S2 acule, !40, 210 adagio, 100 adapt, 92 addict, J14, 172 addicted, 169 Luidilioiial, 2S

adjacent, 2B. 87 adjective, 366 admiral, 8(, 1(11 admit, 114, 157 admonish, 172 adolescence, S2 adventure, 140 advert, i&4 advcrlisement. 140 affability, B2 affect, 193, 322 affection, !72 affirm. 184 afford, 125 affray, S2, 140 ЛГЬСЮ. 141 African. 129 after. 25? afternoon, 127 aftershock. 127 agenda, 257 i. 82 n. 25. 133 a.ghajt, 365 agree, 136 aid. 112.309 aide-de-camp, 116 ain't, 29, 90, 277 air, 82 airgun, 124 jiirport, 12(1

alacrity. 367 alarm, 100 , 101

albino, 42, 101 alchemy, 101, 255 alcohol, 10L alcoholic, Ш , 140 alcoran, 101 alcove. J0I aldehyde. 141 alembic, 101 alfalfa, 101 algebra. 101 alive, 131 alkali, 101 all in, 205 all set, 205 allah, 101 allegation, 36S allege. 36» allegedly, 132 allegory, 87 allegro, 195 alley, 348 alligator, 100 allow, S2, 114, 309 allowance, 114 all own nets, 113 allude, 176 almanac, 101 alone. 19 2 alpaca, 101 aluminum. 44 always. 258 am, 264. 266 amateur. 110 amazement, 133 amaziriply. 191 amazon, 137 ambassador. 19& ambition. 25. Ш amen, 102 amend. [JO amiability, 133 amiable 110, 133 amiabiencss, 133

382

Index of Modern English Words

amok, 102 amongst, 131 amorous, 110 ampere, 137 amused, 125 amusement, 123, 13 I :s. 125 amusins, 125 an, 239-242 analysis, 115.114, 184. 367, 368 analytical, 367 analyze. 112, 1)4, 367, 36S uncephalize, 92 anchorite, 55 anchorman, 121 anchovyL LOO and, 9S andante. 100 andcrson. 136 ал gel, 366 angelic. 366 angleworm, 107 animal. 162, 169, 1*4. 255 announcement, 187 answer, 115, 1-6 anKwcrrd. 222 ant. 46 antagonism, 298 :intc-chamher, 127 aim-aircraft, 127 antimonyц [01 antipodes. 110 anxiety, 88, 89 anxious. 19? any. 112 anybody. 279 anyone, 279 anywhere, 279 apartment. 109 ape ICl apocalypse, 167 apothecary. 82 appear, 2 JO app-asc, 13й appellation, Я2 nprlaud. 224 apple. 320, 331, 337 apply, S2, 112, 140 appointment. 188 apprehend, 17? approach, 82.136 appropriate, 92 apricot, 101, 161 arcade, 1Й0

archangel, 81

arch-thief, 127 are, 264 argot. 20J argue, 172 ana, 100 arm. 135 arrnada, 100 armadillo, 100 army, Ы aroma, 176 arrange, S2 arrest. 172 arrival, 136 arrive, 81, 136 arsenal, 101, 102 art, B2 artichoke, l№ artifacttial, 204 arlillery, 171,205 arltsan, 100 as. 3L0 ascend, 112 a&k. [ 15 p asprc, 210 ass, 337 assailant. 185 assassin, 101, 137 assault, 185 assembly, 32. IS7 assertive, 130 assignation, 188 assist. 112 assistant, 129 assume, 367 assumed, 168 assumption, 367 astonishingly, 191 astronomy, SI asiute, 146 at,219 h 2SS. ЪЪ ate, 219 iithlctc, 114 athletics, 355 atmosphere. 92

atoll, 102, 195

alone, 112 attack, 100 attempt. E15 attend, 140 attire, 114 atlomey, 82, 90 auburn, 161 auctioneer, 129 aunt. 34, 81, 160, 176 austere, 210, 211

Austrian. 130 author, 3I3-J19 authority, S2 auiohypnosis, 127 atilomania, 131 automobile, 141) aversive, 113 avocado, 100, 104 avoid, 136 avow, 136 aware. 125 awful, 207 awfully, 191 awkward, 6\ ак, 203, 302 axed,310 axle, 46 azimuth, 101, 102 azure, 161 babble, IS babel, LOZ. 137 babushka, 101 baby bujjsy* 107 baby tab, 107 haby carriage, 107 baby coach. 107 babysit, 89. 123, 134, 13S babysitter, 134 back, 317 backbite, 134, 139 backbiting 134, 139 backpeddlc. [U backup crew, 121 backward, 96 backwards. 96, 258 bacon. 90 bad, 347, 358 badger, 55 baggage. [09, 129, 196 bagpipe. 120 baH, 82

bait, M4. 311 bake, 261,295, 317 baker, 90 balcony, 100, 205 haJk, 3"l8, 362 balloon, 100 bells, 206 balm, 362 bamboo, 46. 102 banana, 100 band, 61 bandana, 102 bandit. 100 banditti, 257 bang, 166,310

of Modern English Words banged, 309 twnaLjp. 205

banjo, 109 bankrupt. 100 banshee, 35 banl, 136, 137 banler, 95, 207 baptist. 81 barbecue, 100 barber, 90 bare, 366 bark, tOO, 261 barn. 175, 308, 357. 358, 360 baron. 81, 82 baroque, 100 barrack, 100 barrac uda, 100 barrel, 311 Ьм.чш, £1 bass, 256 bat. 144, L84, 30*, 310 batch, 295 bate. 311 hath. 337 bathe. 310. 337 bathroom bowl, 187 ba!hs. 309 battalion, 100 battle, 50, K2 bawd. i96, 310 bav.1] out, 114 bazaar. 102, 116 be, 67, (08, 115, 163, 230, 259. 265, 267274, 282-283 beacon, ><>5 bead, 187, 188 benn, 48 bear, 4(5, 26 К 320, 346, 366 beast, 174 beat, 261, Hi beatable. 132 beautician, 130 beautiful, :&, 132 beauty. S2. 135 beaver, 40 became. 219 beckon. 295 become. 218. 219,230 bed, 325, 547 bedeck, 127 bedlam, ]3(>, 137. 360 bedouin, lrji bedtime. 120 beech. 46

beef, 82, 90, 205 been, 30, 277 beer, 116 beet, 311 beetle, 337. 339 befall. [35 befriend, 127 Ьееваг, 133 begin, 261 behalf. 323 behave, 114 behavior, 114 behavior ist, 130 behemoth, 102 behind, 2Я being, 266 belabor, 166 belch, 98 belcher. 136 belfry, 192 believable, I!5 bell, 55 belly-bumper, 107 belly-buster, 107 belly button. 201 belly-flopper, 107 belly-slammer, 107 bell. 51 bench, 33 bend. 176 benefit, 92 benzedrine. 140 Bertram, 302. 303 beside, 28 bestride, 261 bet, 144, 309. 311 betray. ! 14 between. 2S, 126 between times. 139 hcwilth: 184 biddy. 197 big, 24, 146. 209, 329 big shot. 205 bigger, 233 biggest, 233 bijjmyulb, 39

bigmouthed. 39 brunt, 207 bikini, 137 bile, 366, 167 bilious, 366. 367 bill, 114,257,310 bill, 219 bin. 51 bind. 261 birch, 46, 6t

3&S

bird, [62. [75, 176, 219. 341 birds, 312 bit, 309, 310, 311 bituh, 197 bite, 135, 261, 323 btller, 184,209,210,363 bi-valve, 127 blab, U3 blackguard, 197 blackout, 123 bladder. 337 blame, 157, 336, 34J, 346 bland, 209 blarney, 55, 116, 137 blast, 135, 1K4 blaxploilation, 141 bleachers, 205 bleat. 143 blend. 204. 261 blender. 42 blight, 207 blind, 343 bliss, 295 blithe, 295 blitzkrieg, 11* blobstetrician, 141 blood, 344 bloodhound, 120 bloomers, 137 blossom, 199, 201 blow, 261 bloi*1 up, 184 blue, «2, 99 bluffs, 205 blunderbuss, lOi blunt. 208, 211 blurb, 205 blurt, 143 BM, 199, 202 BO, 202 boar, 90, 162 board, 174,265 boat, 144, 309, 311. 326, 335, 336, 344 Bobby, П0 bog. 55 boil, 82 boldness, 124 bolero, 101 bolshevik, 101 bolster. 207 bomb, 36Й bombard, 36B bombed, 262 bonanza, Kll bone, ЗЛft, 344

:iS4

Index of Modern English Wo ids

bong, 143 bonk, 143 book,4fi, 156, 239, 257, 344 book bind. 134 hooker^!, 122 bookmaker, 140 books, 219 bookseller. 139 boom, 42, 143, 293 boomerang, 102 boon, й 1 Ьоет, 101, 197 booster, 205 boot. 82, 311,325 booth. 6 I booties. 134 bootlicl;.. 134 booze, I0I, 207 bordello, 196 borough. 48, 50 boss, [01.205 both, 61. 321 boiher, 311 bottle, 517 bottom, 311 bottomed. 312 bouBh. 310, 366 bought, 37, 144,223. 30*, 310. 326 bounce, 101,114 bound, 336 bourbon, 1 37 bow, 329, 366 bowdlerize, Ш , K»5 bowler, 137 bo*. 176 boy, 24, 25. 35. 36, 101, 144, 156. 162, 15*7, 23S, 308, 305», 310 boy friend, 120 boycott, 138 boy's, 233 bra, IBS bras, 157 brain, 226 brainstorm, 205 brainwash, 134 brandy, 101 brand у wine, 140 brass, 295 bravo, 100 brawn, 90 braiftn, 295 breach, 29 5 bread, 1 «4,319

break, 223, Ш , 295, 319, 346 break wind. 201 breakable, 129 breakfast, 140-141 breaxl meat, Ш6 breastfeed, 134 breath. 319 breathable, 132 breccia, 100 breed, 295 brethren, 257 brew, 4E. 261 bribery, 130 bricklayer, 134 bride, 326 bridge, 199 brief, 3?, &2 brigand, 100 bright, I 14, 146. 1S4, 196,209.341 brilliant, 209 brisk, 211 broad, 197, 209 broadcast, 195 broke, 205. 223 brokenhearted. 133 bronco, 101 bronze, 46, 100 brooch, 32$ brood, 295. 344 broth, 48, 337 brothel, 196 brother. 22, 34,46, 160, 257, 320, 344, 359, 360 brow, 1Я4 brown. S2 bruise. 135 brummagem, 136-137 brunch, 140, 141 brush, 135 bucket, 82, 107 buckle, 135 buddy. 205 budsc. 166 buffalo. 256 his, 109 build, 135. 317 building, 187, 265 building engineer, 199 built, 265 bull. 61. 106, 162 bully, 95 bulwark, 101 bum, [01 bump, 143. 147 bungalow, 102

bunk,137 bunxen, 137 buoy, 10! burden, 135,34] bureau, 170 burglar, 133 burglary, 114 burke, 136 burlesque, I(K) bwn, 26-1 burr, 311 burro, 101 burst, 261 Ьич, 207, 303 biiih, 303 but. 93. 309, 311 butcher, 90, 176 butler. SI, 9Q buit in, 205 buster, 50,302, 331 button, 82, 302, 311 buy, 37. 48. 50, 112, 310. 366 buzz, 143 hu7.jjed, 2fi2

BVD. 202 by, 2B. 219. 309, 366 cab, 42. 207 cabal. 102 caboose. 101 cabriolet, [JO cad, 207, 326 caddy, 102 cafeteria, 101 cairn, 55 cake, 226, 321,346 calendar, 82 calf, 90, 32«, 336, 349 calico, 102, 137 call, 61, ЗОВ, 359 called, 262 Calypso. 167 са!ул, 167 came, 346 uamel, 46 camera, 110 camouflage, 205 camphor, 101 и:т. 2R, 186, 267, 26S cancer, 198, 200 candle, 51 candy, 42, 101 caniculture, 138 caninal, 13B caninity, 138 canned, 310

Index of Modern English Words cannibal. 100 canoe. 104 canon, SI cant, 204 cantata. 100 canter, 1Э6, 138 canticle, til canto, 100 cunvastack, 1<52 canyon, 101 cap, 302 capital punishment, 199, ZOO capitulate, 114 caprine, 100 captain, 82, [24 captaincy, 129 car, 226, 357, Э 5 8, 360 carat, 100 caravan. 102, J40 carcarricrs. 298 cardiac, 321 cardigan, 137 cardinal, Bl. Я2 care, 321, J37 cargo. 100 caribou, 256 carmine, 161 carnival, 100 L-iirnu5c. 101

сигр, 172 icarpenler, 90 carry, S2, 176, T77 cars, 119 curl. 348 cartoon, 100 Carve, 261 carves, 224 carving. 131 .че, 323 cashmere, 102, 137 casijio, 100 east, fftO cassava, 1(14 cast, € | casrle. Si]. 255 castles in Spain, 116

Casual, 5* casualty, 199 tat, 184. ЗОЯ, 315, 31ft, 326, 328, 33(5, 348 catacomb, 100 catalogue, 36 s catch, St, 32S catch on, 205 caucus, 104

caul, 167 caused, ~^4 cause, 224 caustic, 92 cavalier. 100 ceilinj;, 82 cell, «1, 167,316 « l i a r , 167 cellule, 167 cement, 135 censure, 172 center. 364 central city. 199 centre, 364 cents. 310 centurion, R8 Leph.ilopod, 110 cereal, 137 certain, 82 certainly, 132 chaff, 328 chainsmoke, 134 chair, 33, 35, SI. 174 chalice. !67 chalk, 50, 135 chalked, 310 chamber. 110 champaign, 137 chancellor, SI chandler, 90 change, 81, 82 chant, 203 chaos. 28, 92 chap, I 14, 140 chapel, 81 chaplain. 82 chapter, 81 characters, 82 charge, 205 chartreuse, [61 chaste, 1 И , 366, 367 chastity, 366, 367 chat, 38 chaltefboA, 39 chautauqua, 104 chajvanism, 137 chauvinism, 195 check. 364 trietked. 310 cheukmate. 102 cheek, 328 cheery. 166 cheese, 46, 50 chemical substance, 174 cheque. 364 cherry, 82, Ifil cherubim, 101, 257

385

chess, 82, 102, 116 chest, 209 chestnut, 161 Chevrolet, 140 chew, 115, 135,261 chewed, 260 chiaroscuro, 100 Chitsigo-lype, i j ] chicten, 90, 162, 176, 239, 328 chicken less. [06 chicken out, 114 chide, I 14, 135 chides, 311 chief, 2B chij-acr. 109 child, 135, 2?7. 336 children, 160, liJU. 257 children's, 233 chili, 104,226 chill, 135,328 chimney. 82 chin, 309, 328 china, 137 chinchilla, 100 think, 309 c h i n b , 310 chintz, 102 chintzy, 207 chip, 352 chipmunk, 104 chippy. 197 chit, 102, 196 chocolate, 42, 104, 161, 32 S choice, 3 11 choir. 317 choke, 32 S choose. 135, 261, 318 chortle. 140, 141 chnrus, 302 chose, 302 chosen, 318 chow, 205 chrism, 112 Christ, 112, 153. 201-202 Christianity. 130 Christinas, 3fi2 chronology, 92 chuck, 205 chuckle, 141 church, [35 churl, 197 churn. [35 chutney. 102 cigar, 101 cigarette, 101, 365

386

indc* of Modern English Words

cilia. 167 cinch, 101 cinema, 114 cinnamon, 102 cipher, 101 circulate, 221 circumcision, 81 circumnavisate, 127 citadel, loo cite, 366 citrine, 1Й1 city, 81, Я2, 316 civil, 367 civility, 367 clLibber cheese, 107 Clack, 38 clan, 55 clandestine. 167 clams, 42. 120. 143 clank, 143 claptrap, 99 clasp. ПО classy, 99 clatter, 143 clean, 224 clear, 82. 209 cleared. 311 cleave, 261 clergy, В2 clerk, SI clever, 146, LSI, 166 cliffs, 48, 325. 32Й climb, 112, 261, J36 Cling, 261 clip, 61. 135,204 clique, 302 cloak, 54 clo£, 110 dose, 82 close up, 205 clothes. 114 cloud, 135 Clown, 197 cloy, 210 cloying, 109 club, 321 clucked, 310 clump, 147 clunk, 143. 147 coach. 329 coal mine, 120 coakmoke, 124 coaly ard, 124 coarse, 197, 209, 211 coal. «2 coax, 166 cobalt, 101

cobbler, ПО cocaine, 140 cock, 201 cockatoo, 102, 116 cockroach, 100 cucklail hour, 109 coconut. 195 cocoon, 104 cod, Ш , 358 co-education, 140 coffee, 101. 102 coffee break, 10? cog, _32<> cognition, 89 cognizance. S9, 133 cognizani, 89 cognoscenti, 89 cohort, 82 coincidence, 25 cold, 44. !56, 1 R 209, 210, 29?, 320 collene. 55 colloquial, 98 cologne, 137 colonnade, 100 color. 82. 167,364 colorfasi. 126 colour. 364 comb, 321, 336 combat. 82 combustion, S2, B9 come, 261 comet, 90 comfort station. 199 comicstrip, 120 command, 349 commando, 101, 112 commence. 223 com mend. 15$ commode, 186 common, £2. 192 commonly, 113 communist, 140 compare, 19? compendious, SB complain, 136 complement, 322 compliment. 322 compose, I £4 comprehend, 175. 179, 184, 196,206.207 comrade, 100, П0 comstoctery, 137 conceal, 167 conceive, 89, 184 concert, 100 concerto, 100

concise, 39 conclude, 184, 196 concrete, !84 condemn, 172, 317 condemnation, 317 condom, 137 condor, 104 cone, 366, 367 confederacy, 129 confess, 114, 157, 224 confessor, БI confidence man, 140 confounded, 191 confused. 114 contusion, 341 conic, 366, 367 conicity, 367 conjecture. S2 connect. 18J connection, 364 connexion, 364 consolidate, 92 conspicuous, 92 conspiracy, 87 constable-, 81, 198 conslcllatioti, 90 contempt, 87. 133 continual, 193 continue. £2, 223 continued, 222 continuous, 193 contraband, 100 contrapuntal, 100 contrary, S2 contrast, 193 conventioneer, 129 conversation, 82 converse. 38, 115 convict, HO cook, 50,219,221,222, 223 cookie, 101 cooks, 219 cool. 184,209,295 coolic> 102 cop, 114 copper, 50, \M, 137 copra, 102 copulate, 201 copyread, 134 cordial, 112 cord wain, 100 CORE, 141 cork, 187 corn. 4S. 171, 328 cornice, 100 correctness, 132

e]f Modern English Weirds corridor, 100 corselcirEum, 131 cost, 82 cut, 102, 316 cottage cheese. 107 ccitton, 101 cougar, 104, 116 cough. 302 could, 267 council. SI count, 82 сои п (down, 121 countenance. I !4 cnunterstalemenl. 127 countess, 8T country, 25, Я2 courage, 321 course. 82, 2 1 0 court, SI courtesan, 197 cousin, J4, 81, 140 a>vcr. a: covet. 169 cow, 46, 90, 162, 257 cow], 16? co-worker, 127 coy. [90 coyole, 104 « и , 140 crabhed, 39 crack, 99, 135 cracked. 185 сгасЫе. 120 craft 337, 34$ crafty. I № crag, 55, J02 cram, 34<J crane, Ifi4 crank, 135 crap, 205 trash, 143 crasher, 205 crave, 61 Crawl, 61 crazy, 185 cream, 82, 112, 153 creation, ]87 «*ed, 116 creep, 135. 261 Creole, 100 crepitate, 200 crescendo, 100 crelin, 112, 153 crinianemlies, 202 crime. 42, S1 i assault, 199, 200

cripes, 201 crisp, 39 critic. 92 criticism, 130 criticize, 157 crone, 197 croak, 16R crooks. 168 crop, 3ZS enow, 2ft 1 crown, 61, 1Я8 crucified, SS crucifix, 81. B2 criid cheese. 107 cruds, 107 cruise, 101 crumb, 365 crumple. 147 crunched- 310 crush. 13.6 crystal iue, 130 cub, 110 cuckold. 197 culturally deprived, 199 cunning, 172, 190 cunt. 166, 201 cup, 51. 337 cupidity, 137 cupola, 100 curb, 364 curd cheese. 107 curds, 107 curie, 137 curricula, 257 curry, 102 curt.' 39 curtain, 82, 135 custody, Й7 custom, S [ customer, 140 cut, 184. 31ft cute, 99 c a r , 101 dachshund, [01 dad-blame, 202 dad-burn, 20 2 daddy, 130 dado, 100 dago, 101 daguerreotype. 13<J dam, 197 damask, 137 dame. 81, 197,336, 345, 346 damn, 172, 309, 317, 36R damnation. 317. Дб£

387

damned, 191 dance, 82, 349, 358 danced, 262 dancin', 2156 dancing, 26й dandle, 110 dang, 202 dangerous. Si darb, 205 dare, 321 dark. 1S4, 209 darken. 130 darn,, 201 darn ins needle, 107 dart, 135 data, 257 date, 183 daughter, 34, 46, 160 davenport, 137 dawdle, 114 dawn, 195 day, 295, 329, 340 days, 131 dead, 323 deadly, 115 deaf mute, 123, 139 deal, 325. 345 dealt. 262 de-Americanize, 126 death, 31У debt. 317 decade, 113 decay, M5 deceit, R2 deceive. 89, 114 decide, 120, 134 decision, 120, 134 decontaminate. 127 decoy, 101 deed, 295, 321,325 deem, 295 deep, 114. 184. 209, 320 deer, 46, 90. 170, 171, 174, 196,256 defeat, R2 defecate, 200 defence, 13ft, 140,364 defend. 139, 140 defendant, 82 defense. 82, 364 defile, 29? define, 184 deft, 321 defuse, 126 deity, 112 delay. 114 delicatessen. 101

3$$

Index of Modern English Words

delineation, 88 deliver, 168 delve, 261 demand, 112 dernend, 130 democracy, 32 den, 135 denizen. 172 denial, 321 dental plates. 199 dentist, 321 dentures. 199 deny, 157 depart, 115 depend. 184 deplore. 176 depose. l?6 depress un*e, 12fi depth, 114 deraign, £2 derby. 137 derrick. 137, 19 5 derringer, J 37 deruncinatc, 92 description, 187 desert, 46, SI deserts, 172 desire, i l l . 174 desisr. 115 desk. 239 desperado, 100 desperately, 207 despite. HA despondency, 88, 89 destroy, 82 detective. 140 determines, 221 detest. 116 developed, 222 devil, 189,359 Jesterily. 92 Diana, 112 diary, 112 die. 61, 112, 175, 222 difference, 135 different, 13 L diffuse. 39 dip. 184, 20*, 529 digg[n£, 331 dillydally, 114 dim, 146, 209 dimension, Й8, 112 dm. 326 dine, 112, 222. 224 dined, 222 drnshy, 102 dingo. 102

dinner, 82, 20^ dirt, 61, 176 dirty. 184 disabled, 199 disagree, 92 disallow, ]27 disappedr, 218, 219, 222 disappeared, 219 discern, 317 discharge, 114, 136 discipline, 81 discovery, 187 disdain. 82 disease, 171 disestablished, 298 disburse. 1315 dish, 51 disintegrated, 222 disinterested, 193 dismal, 112 dismiss. 114 dismissal, 113, 114 disparage, 82 display. [ 12 disport, 140 disposal. 129 dissident, 112 dissolve, 223 distillery, [40 dislract. 87 distribution, 82 dive, 330 diversion, 120 diversity. 82 divert, I 20, 368 divest, 176 divine, 129 divvy up, 205 do, 163. 264. 281,290, 295. 320 doctor, 140 doctrine, 317 dodge, 110 doc., 42. 46, 166, 175, 184,207,239.329 doge, 100 doggish ness, 138 doEEone, 201 dogey- 138 doghouse. 138 dogma. 92 do-Eondcr, 163 dog-raising, L3E dops, 206 t! • •: i i;j -. 1 6 6

doll. 349 donation, 133

done, 277, 344 doodle, 206 doo-doo, 202 doom, 135, 172,207, 295, 323 doomed, 168 Doomsday, 168 door, 46, ^n dope, 99, 101, 146, 153-154, 205 dope-sheet, 154 dopey, 114 dope; people, 185 dorm, 120, 140 dormitory, 120, 140 dol. 302 dotard, 101 dote, 101 double cross, 114 doublepart, 134 doubling, 131 doubtful, 132 dough, 48, 184, 205 down. 61 doiy, 197 dozen, 82 drab, 197 draft, 364 drag, 184 dragon fly, 107 drank, 261 draper, 90 drat, 201 draught. 364 draw. 261. 329, 333 drawed, 261 dread, 135, 261 dreadfully, 191 dream. 193 dreams, 323 drift, 204 drink, 26! drink ability, 132 drip, 184, 295 drive, 177, 261 driver, 130 drives, 2L9, 224 droop. 61,62, 304 drnp, 295 droshky, 101 druggisl. 109 drurnslicks, 1Q<5 drunk, 261 dry, 209 dry clean, 134 duuu, 100 duct, 162

\ndcx Ы Modern English Words dude. 205, 207 dug, 147, 261

dukc.ei, 82

dulcet, 210, 211 dull, US. 208, 20?, 210, 211 dumb. 99, 114, 207 dumdum, 102 tlu m found, [6f> dump, J43. 146, 147 dumps, 110 dunce, 137 dungaree, 102 dupe. 114 dust. 224, 226 Dutch cheese, 107 dwarf. 348 dwell. 4!, 193 dwindle, 304 eager, В2Л93. 210, 211 eagle, 46, 162 carl, 48, .10 earls, 312 earth. 336 earth-born, 122 earthly, 258 earthworm. 107 earthy. 130 ease, $1, ?23 easel. 101 CA5t*4S easy, 82. 121 easy j-ainj, 138 easy-going, I?? easypoing. 123 eat, 41,44, 112, 115,203, 220, 22[.224, 261, 309, 323 ecstasy, 1 12, 172 edge, 135, 352 editor, 133 education, 92 education wise, 131 eel, ЗОВ effect, 193, 32? Bffigy, 172 e£fi, 61 egghead. 99 efis-hr;id, 163 s, 219 h. 302, 303 r. 125, 127. N0 elastic, Збй elasticity, 366 l l , ]36 ric. 368

electricity, 368 electrocution, 133 elephant, 46 rlephant дип, J24 eleven, 112 eliminate, 114 eliminating, ] 13, 114 elixir. 101 elliptical, 39 elm, 4* elope, 176 emacerale, 92 emhargc, I0A embolden, 127 emerald, 161 emigrant, 322 emir, 101 emotion, [33 emphasis. 92 employ, ] [2 employed, [[3 ещпгезь, 81 empty, 341 encircle, 127 encounter, 82 encyclopedias. 29& end. 337 ended, 262 enemy, 82, 110 enervate. 112 enforce, Г 36 engagement, 188 engineer, 112, 19R engrave, 261 enormously, 191 enough, 127 enroled. 364 enrolled, 364 enter, 136 entered, 125 entering. 125 enicrment, 125 enlers, 125 enthrall. 134 enthuse, 94 enthusiasm, 92, 133, 137 entrance, 125, 129 entymolo^y, 29B envy. 140 epigram a tic, 39 erotic, [3Sr 172 eructiiie, 200 escape, 136 escheat, 140 cspicc, 14(1 espy, 140

estate, 140 esteem, 136, ]72 estrange. 140 ether, 332 etiquette, 42, 140 euphemism, 198 even, 209 ever, 3|6 everyone. 127 evidence, 82 evil, 326 examination, 140 example, 140 excavate. 115 except, 322, 359 exceptional child, ]99 excessively. 19J evciccate, 92 excise, Id I cAcl;imation. 133 excuse, 82 execute. 171-172 exercise, 135 eximiuus. $2 exist. 92, 184, 219, 220, 222 existence, 129 exisl.4, 2)9 cxpectafion. <)2 expectorate. 199, 200 expedienl, ] 1Q expedile, 110 expedition. 110 expensive. 92 experiment, 135 experimental, 1 [3 experimenicd> [ 13 expire, 112, 199,200 explain, 92, 184, 196. 206. 207, 224. 225, 367 explanation, 187, 367 explanatory, 367 explicate, 112 explicit, 82 explode, 165-166 evpnrt, 112 external, 92 extraordinarily, 191 extraordinary, 95 extrasensory, 298 extra-terrestrial, 127 e x t r a v j i g a n l l y , I*)]

extremely. 191 extremism, 130 exult, 185

390

Index of Modern English Words

exulting, Й8, 89 ci-wife. \.?7 eye, 135, IB4 eye of a needle, 207 eyes, 20tf fabulously. 191 face. 81, 114 fact, 176, 177 fadaise, 140 fade, 205. 346 fade out, 20 S fahrenhcit, 137 fail, 203 faint, 168. 209 fair, 20? faithfulness, 132 fake, 205 fakir, IOI fall, 259, 2M>, 26 L, 348 fallen. 260, 261 falter, 34K fame, 172. 198,207,336. 346 fan, 114,206 fanatic, 140 fanny, 199, 202 fantastically, l^ 1 far, 359 farad. 137 fare, 261 farm, 135,348 farmer, 90 fart, 99, 135 fascinate. IS4 fashion, 82 fast, 156 fat. 321 fate. 137 father, 34.41,46, 112, 160, 30S, 309,320 father-in-law, 34 fathomed, 312 falmouth. 39 fa!mouthed. 39 fault, 82 fawn, 161, 302 feast, S2 feature, 204 fed, 294 fee. 320 feed, 294,295, 321 fee!, 135.230,294,323 feet, 144, 206, 257 feldspar, 101 fell, 259, 260. 261 fellow, 61, 114, 135

felt, 223, 294 feminine hygiene spray, 198 fend, 139 fetlock, 110 fetter, L10, 111 fever. 51 fez, 102, 116 fiasco, IW field, 50, 336 fiend. П4, 173 fig. 329 fight, 224, 261, 321 figuraiive, £8 figure, 82 figurine, LOO fill, 295. 309, Ш films, 114 filth, L35. 163. 173 final, 82. 255 final solution. 199 finale, 100 finalize. 130 finch. 48 find, 2(9.220. 261, 336 finding. 187 finds.. 221 fine. 82, 135 finesse, 207 finger, L «4,321 finger of land, 207 fire, 114, 131, 2.07, 326, 343 firebLrth, 138 fire bus, W firefishler, 134 fi rehouse. 124 fireman, )24 fireplug, 124 firkin, 101 firm, КЮ firs, 312 first. 359 fish. 45. 48, 49, 256, 257, 320 fish bait, 107 fisherman, 90 fishery, 130

fishfurter, 130 fishing worm, 107 fishmonger, 50 fishwife, 34 fist, 135,326 fit, 144 filled. 262, 312, ЗЙ7

live, 320 fix, 199

fii up, 239 fixed, 219 flaccid. 147 flack, 141 flail, 147 flame, 147, 345 flange, 309 flap, 147 flare. 147 flash, 143, [47 flat. 61, 147, 184, 209 flatulent, 147 flaunt. 147, 193 flavor, 135 flay, 2Й1 flea, 147 flee, 147,261 fleet. SI. 147. 295 flick, 143, 147 flicker, 147 flicks, 114 flimsy, 143, 207 fling, 147 flint, 147 flip, 143, 147 flippant. 207 flirt, 205 flit, 61, 147 float, 147. 295. 34J flog, 147 flood, 320, 344 floor. 139, 147 flourmverings, 298 floozy, 197 flop. 99, 147 florin. 100, 116 flotilla, 101 flounce. 147 flour, 147 flourish, 13Л, 147 flout, 147, 193 flow, 147, 261 flower, 82, 147 flower bed. 124 flowers, 223 flown. 260 fluent, 147 fluid, 147 fluke, 99 flume. 141 flunk, 203 flunkey, 147 fluiry, 147 flush, 147 uuslei'. 147 flute. 147 fl ulter, 147

of Modern EngM^b Words flux. 147 fly- 120,147,260,261, 289 dyer, 239 flying mathine, 120 foci, 257 foe, 257 fog, 120, 140 folk, 74 follow, 329 follow-up story, 121 food, 44Л 14. 203, 295, 309, 344 fool, 1 U foolish, 146 toot, 41, 45, 110, 111, 135, 184.257, 320 fop,, 207 for, 98, IIS for crying out loud, 202 for Pete's sake, 202 for the love of Mike, 202 forbear, 125 forbid. 125 force, 25, 82 forcefecd, 134 foreign toundy, 127 foreshadow, 127 forester, 90 forget, 125 forgo, 125 forfc,,9G, 203 forlorn, 112, 125, 321 form, 82 formulae, 257 fornicate, 201 forsake, 125, 261 forswear, 125, 127 forthcoming, 127 fortuitous, 193 fortunate, 1*3 fortunately, 132 fortune, 137. 173, 198 foul, 295, 323, 325 found, 223, 34J fourfold, 130 four-] egged. 125 fiCwt, 17]. 175 fowler, 329 fracas, 100 fragmentary, 129, 207 fr&nfcjurrcr, 101, 136, 137 franiic, 176 freaked out, 207 freckle. 61, 62 freedom, 1 ]fi freeze. 135. 2fil

freezing told, 46 frenzy, 17ft fresco, 100 fret, 26 1 friar, 82 fried, 223 friendless, 130 friendly, 258 friendship,

]3l

frigate, 100 frightfully, 191 frisk, 206 frolEc, 205 froie, 161 frozen, 261 fruit, 81. 82 fruitful, 132 frump, 147, 196 frustrate, £7 fry, 82 frying pan, J07 fuck, 168, 201 fugue, 10» fulfill, 127 full, 191, 193, 209. 295,

m

fulsome, 193 fumble. 147 fun. 166, 207 fund ion, Ы, 92 funeral home, 199 funk. 147 fur, 62. 187 furlough, 101 furnish, 136 furrow, 46, 48 furtive, 82 fury, 137 fuss, 143. 166 fuzz, 205 fuzzy, 184 gab, 38 gabble, 33 gabby, 39 gabfest, [30 fad, 201 gadget, 110, 205 &£, 143 gain, 257 gainsay, 127 Bait, 61 salary, 90 sal I eon, 100 galley. ВI galore. 55 gamble, 207

gambler, [ft6 gams, 206 gang. 176, 177 gap, 61 e, 116, 351 gar Jen, 116 gargantuan, 138 garment, 199 garrulous, 39 gas, 184 gasp, 3115 gater 321 :• : : S - \

•

I " '

gaudy, 190 gangs, 317 gazelle, 101, 102 GD, 202 gee whiz, 202 geese, 257 geisha, 102, 116 pern, 51. 309, 316 general, 112 geneva, 140 genie, 101 genitals, ) 12 genius, E7, 112, 137, ]4ft genre, 35 [ gent, 140 gentleman, 112, H4, 140 gentleness. 132 George, 219 gerrymander, 138, 140 gestapo, 141. 148, \95 gesture. Я7 get, 61. [84, 196,261,32? ghetto, 42, 100, 116, 195 ghost, 168, 173, 174, 317, 321,329 ghostwrite, 134 ghoul, 101 gift, ^42 gilded youth, 11 ft gin, 101, 136, 137,316 ginger, 102 gingham, 102 ginfco, 329 giraffe. 101 girdle, 200 girl, 35, 3ft, 157, 162, 197, 226, 239 girlish, 130 give, 61, 261. 329 give up, 114 Elsce, 147 glacier, 33, M7 glad, 147, 329

2&2

Index nf Modern English

glade, 147 jl:i.li.:.lu!. 147 gladioli, 257 gladiolus, 147 gladly. 337 ::..

i : ••!• . 1 4 7

glance. 147 (land, 147 glare, 147 glass, 147. 187, 295, 337, 349 glaucoma, 147 glaze, 147.295 glazier, 90 gkam, 135 glean, 147 flee, 147 glen, 147 glib. 39. 147 glide, 147.260,261 glimmer. 147 glimpse. 147. 317 glimpsed. J09, 334 glim, 147 glissade. 147 plisfcn, 147 f i l l e r . 147 gloaming, 147 float. 147 globe, 147

gloom, 147 glory, 5S7 147 gloss, 147 glottis, 147 glove, 147, 344 glcw, 147. 199,201,329 glowing. 146 glue, 147 glum. 147 glum p. ] 47 glut, 147 glutton, fti glycerine, 147 gnat. 329 Sn:LlK, 310 gnaw, 261 gneiss. 101 gnomic, 317 go. 44, [76. 177, 259,261, 264. 355 so to bed with, 20 [ go to Ihc bathroom, 199 Boal, 364 goal, 4Й, 257, 329, 344 ЕоЫсТ. B2. 90 god, 337 God, 201-102

goddess, 130 godlike, 130 going. 121 gol durncd. 202 gold, 46, 48. 50. 329 golly, 201 gondola, 100 gong, 102 goober, 109 good, 34, 44, 219, 329, 344 goodby, 27 goof, 205, 207 goose, 257, 321, 344 gosh, 201 gossip. 38, 39, 135. 196 gossipy. 39 got, 316 governmeiH, 82 gown, 32 grab. I «4 grace, 137 graceful, 132 grade, 329 graduate, 140, 362 graduation, 18-6 grafl, 206 grain, 46. 43 gram, 364 grammar. 82 grammar sthool, 120 grammatical, 204 gramme. 364 grandly, 191 grasp. 170, 178, 179-1 SO. 181. 185, 196, 206, 207 grass, 34, 196. 239, 295, 341 grate, Я2 graveyard, 201 gray, 3(>4 gra/jc, 295 greal, 319,346 greatly, 191 green, 120 green bi:an*i, [07 greenhouse, 120 grenade. 100 grew, 214 grey, 364 grief, 82 cricf lhcr;ipisl, 199, 201 grirL. 190 grind, П 2 . 135, 153, 261, 320 groove, 101, IK4 groovy. 173

grope, 135 groito. 100 ground. JiU ground meal, 120 grow, 222,223-224. 261, E » s , 222 growth. 131, 19ft, 199. 200 b, 114 це, 173 grummy. 207 grunip, 120 guard, 317 KuerrillEi, 101 gjess, 61, 116, 329 guest, 135. 317 guidance. 136 guise. IU7 guitar. 100 gull, 55 gulp, 143 Bum. 82, 316 gumbo. 109 gun play, 168 gunk, 147 gunny. 102 guru, 102 guih, 143 gats, 184 gutlu-percha, 102 fiuy. 114. IH, 137 eym, 309 gym teacher, [20 gymnasium. 140 gyp- 309 haberdasher, 90 hacienda, 101 buckney, 140 had, 268 haddock, 256 hag. 197, }2<J hah, 143 hairpiece, 199, 200 hairs, 323 halberd. 167 hale, 318 half. 353 bnlitosis, 200 hallelujah, 102 hallow, 295 ham, 316, 325. 336 hamburger, 101, 136, 137 hummer, 33, 135 hammock, 104 hamster. Id I

Index of Modem English Words hand. 41, 44. 49, 135, 336 handiwork, 125 handpick. 134. 139 handpitkcd, 1 ЗУ hands. 184. 306 hanHsew, 134 bane. 184,2ft 1.309 hanged. 96 hangnail. 192 hanker. 101. |66 Hansen'H disease, 199 hansom, 136 happiness, 124 hara-kiri. 102 harangue, N2 hard, 156, 20°, 2Г4 harem, 1I>1 harlm, 196 harm, 34Я Ьагтолу, 137 harpy. 39. 112. 197 h i r r i d j n , 197 harsh, 209, 211 harvest, L76 has, 348 hashish, 101 hassock, 33 hal, JOB, 348, 349 hatched. 3U9 bate, 32.1 hatred. M l have, 25, 30, Л7. 135, 139, 163. 264, 267274, ,120 have to go, 199 hawk. 46, 321. 341 hawker. 133 hazard, 101 , | hazel, 161 he. 219, 234, 235, 23B. 247, 344 head, 41, 114. 135, 184. 1*6,319,320, 333, 341 headfast, 130 heal. 321, 323 healthy. 131 hearse, L86 hcan. 3 3 , 4 ] . 45 184, 32(1. 321 heari condition, 199 beany, 112 heat, 30S. 345 heave. 261

heavenly, 25 К heavy, 1Я4, 2(14, 219 he-cow, 106 hecior, 13$ heir. K! belicqpter, 140 hello, li helinei, 16"? help, 112,222. : 6 | HF.[ P. [42 helped. 222 hemp. Ш2, 320 hen, 197 henchman, I9& herinii. 101 henpetk. 134 henpt-tked. 134 her. 234, 23S, 247. 312 herb, Я2 herculean, 13Я herd. « here, 24Я hero, 137 hers, 2Л8, 247 herself, 3S hesitancy, 129 hew, 261 hex. 101 hi, 27 hiccough. 302 hiekey, 199 hickory, 104 hicky. 202 hid. 294 hide. 135. 144, 294, 326 hideously, 191 M«h, I S3, 184,209. 210. 317 hike, 205 hill. 54 him, 234, 235, 238, 247 hinge. %№ hip. 206 nipt. 205 his, 234, 235, 238. 247 hiss, 42, 143 hissed, 262 history, 87, 140 hitchhike, 134 hive, 302 hmmnnmn. 143 bobble, 101 hubby. 321 hod, 310, 326 hoe. 349 hcigde. 134 hi>ho, 143

393

hoist, m i hold, 261 holdup, 12 L, 136 holiday, 17ft holy,29s, 321,329 home, !K4, 31fi, 325, 336, 344 homely. 254 homeward. 131 homewards, 25K homitide, 87 hominy, 104 h h Lines t, 82 honesty. 2? honor, 364 honour, 364 hoodlum, 205 lioodno voodtio. 109 hoodwink, ] 14 hook. 135 hooch. 32Й hop. 101 hope, 321,344 HOPE, 142 horde, (02. 116 horn. 135 horribly. 191 horridly. 191 hors d'oeuvre. 116 ho™:, if>. 120. 311 horsefly. I2il. 123. 124 horsepislol, 124 horsejihoc. 124 horsewhip, 120 hosanna, 102 hose, 101 hosier, 9(1 hoit, 90 hot. 184. 209, 308,310. 326. 344. 358 hotdoe. 138 hotel, [41 hound, 171. 175, 196 hour. 82, 3 17 hoLtri, 101 bouse. 44. 120, 304, 337 hcusebrenk, 134 housebreaker, 134 housekeep, 134 housepftiot, 134 how, 237. JK3 howf, 14.1 hoyden. 14 ft huckster, 101 HUD. 141 bii£. 110, 147

394

Index of Modern English Words

hugely, 191 huEKci'-iTiLijjjjcr, 99

huh, [43 hum, 143 human waste. 199, 200 humble: 147 humble pie, 192 hump. 147, 320 hunchback. 123 hundred, 44, J5.320 bunu, 96 hunger, 321 hunt, 135, 221 hunter, W hunts. 219 hurrah. 143 hussy. [97 hustle. 10] hni. Ш hymn. 317 hymnal. 317 hyperactive, 127 I. 234, 247 iceberg. 101 iden, 34, 239 id mm, 88 idiot. [97 idiotic. 14*

if, 238

i^nftramti'i, 166 iguana, 100, 104 ill, 61, 16& ilkcetirous, 92 itlcjjilimaie. 128 illumination. 115 illustration. IS7 image, 8! imhrcilic, !46 imhrdulii]. 100 iuiiutc, 25. LEI immediately. 208 immensely. 191 immensity. 82 immigrant, 322 immoderately, 191 immoral, 173 immune, 87 impeach, 1 I ft impediment, 11 ft impersonal, I2S implying, [93 imprecation, ПЗ impress, 368 impression, 36B improve, 225 improvement, 136

improves, 224 in. 28, 44, IIS, 219 incarnate, 87 Inch, 112 include, S7 incommunicado, [01 increase, 82, 136, 223 incredible, 87 incredibly, 191 incubus, В7 incumbent. &7 incursion. 133, 198 indecent, 98 indent. 321 indeji, 57 indFjnily, SS indigo, 161 individual, Я7, 131 infancy, 87 infant. 3*6 infanticide. 36G-367 infanticipatniE, 141 irrfer, 99 inferior. S7 inferring, 19J infmile, 317 infleclicin, 364 inflexion. 364 influenza, 100.140 inform, 82. 116 informal, 9Я infraction, 133 ingenious. 82 in-going. 111 iniquity, 58 injure, [ 15 injury, 58 innate, 87 innumerable, 87 inquisition. 133 insane, 92 inKert, 120, Ш insertion, 120, 368 inside. [27 insider, 205 insincere, 128 insalcnl, [85 instruct, 112 insult, 185 insuperable, 133 insurrection, 1J3 intaglio, MO intellect, B7 intellectual, 163 intelligence, BS, 89 intelligent. 114, 146, 184, 219

Intelligentsia, 101, 116 intense, 115 intensely, 191 inter, 199 international, 126, 127 inte.rperit)d, 1 39 interrupt, 87 intraparty, 127 invasion, 341 invention, 187 iron, 46, 187 ironing board. 138 irregardtess, 99 irregularity, 199 irresolute, 112 is. 264, 266, 267, 268. 309 island. 48, 365 Israeiite, 130 issue. 135, 351 h\234, 23&.247 it go« without saying, 116 its, 2.H 238. 247 ivories, 205, 206 jabber, 3B, 143 Jack, 219 jade, 101, 196 jail, 364 Japanese, 110 jar, 101 jai'gon, 204 jasmin, №2 jasper, ][)2 jawed, -lift jazz, 42 7 109, 184 jealous, 81 jc-had, 101 icily, 133 Jerk, 143 Jet, 120, 195 jet-propelled aircraft, 120 jewellery, 364 jib, 101 jifi, 329 jimminy cricket, 202 jinx, 99 jive, 184 job, 110, 143, i66, 207 job action, 199 jock. 114 join, 82, 112 JOIN. 142 joked, 3 1 [ jostle, [ 12 journal, 112 journey, 81, 176 jovial, [ [ 2 . 1 3 8

Index of Modern English Words joy, 81. 302 joyful, 146 joyridc, 134 joyriding, 134 jubilee, 102 judas, 13 Л judge, 82, 317 judged, 113,262

kind, 336 kindergarten, 101 kindle, 61 king. 46, 48, 50, 188, 326 kingdom, 54 kingly. 130 kink, 32g Kiplmesquc, 130 kipper, 32 Я

judgment, 364 judo, 102 jug, 147

kiss, 326, 32 Я k i t , 101, 32Я kitchen, 5 J , 3 2 K

juke, (09 jump, 25, 143 jumped. 317, 361 jumpin', 266 jumping, 266 jungle, 42, 102 jumper, 87 junk, 102, 147, 207 junker, ]()[ juntn, 112 jury, Й2 just, [24, 31Л, 35$, 367 justed, 124 justice. 81 juslified, 124 fussily, 124 jute, 54. 102

kith, 99, 328 K K K , 141 knapsack, 101 knave. 197 knead. 48, 261 kner, 320 kneel, 61, 222 knife, 90 knigbt, 198, 316, 366 knit. 295 knock. 99. 143, 348 knot, 135, 295, 366 knout, 101 know, 74, 75, 199,261, 317 Kodak, 120 koiher, 102, \U kowtow. 102, 116 kvass, 101

kale, 205 kamikaze, [02 kangaroo, 102 karate, Ю2 Katie, 130 fcaytik, 104 keelhaul, 101 keen, 55, 146. IS4, 208, 209, 211 keep, ]35, 168, 328 keeps, 221

k*e, ж kelp, 32 R kerb, 3 M ketchup, 102, 116, 137 kettle, 328 Wiaki, 102 kibbutz, 102, 116 kid, 61, 99, 302 kidnap, 114 Kilchrisr. 167 Kilkenny, 167 kill, I 10, 221 Kilpatrick, 167 kin, 320, 323

labor, 82, 112,364 laboratory, 140 labour, 364 lace, 82 laconic. 39 lad, 110 ladder, 331 lady of the evening, 199 lag, 329 Jager. !01 lagoon, 100 laissez-faire, 116 lake, 55 lamb,"??, 185.317

lamr, J45, 346 lament, 133 lamentation, 133 lamp, 81, 82 lands sake, 202 languor, 82 lap, 135 lapidary, Й7 lapsed, JQ9 larceny, 114 large, 8 1 . 82

.395

largo. tOO lariat, 101 laser, 141 lasso, 101 late, 346 latrine, 255 laugh, 261, 34Я launch, 102, 136 launchpad, 121 laundreteria, 131 lava, 100 lavatory, 186 lavender, 161 law, 329 lay, 96, 197 lay up. 136 lead, 4$, 50, IB7, 262 leak, 205, 2O<5 leap, 261, 345 Irarn, 295,317 leather, 317. 365 leave. 224, 294 lehenstraujll, 116 led, 262 leech, 48 leer, 173 left. 224, 294 leg, 61. 329 cjiU. 87 d> 129 legs. I7S, 206 leisure, 82. 302 leitmotiv, 101 lemon. 101, 16t length, 131, 135 lengthwise. 115 leprechaun, 55 кккоп, 81, 331 lethal. 138 letter, 135 ieuuee, $2 leukemia, 112 level, 209 leveled, 364 leviathan, [02 levis. 137 lewd. I<J7 Lewisite, 136, 148 liberty, 135 library, 355 lice, 257 licence, 36J license, 113, 135,364 lie, 96, 261, 366 lift, Si, 114 light, 112, 121, 135,209, 335

Index of Modern English Words like, 135 lilac, 102 lilies, 205, 206 Lilliputian, 136 limb. 30$, 325 limbo, 87 lime, 100, 324, 325 line, 5 b 82 Linen. 1Я7 link, 61 lion, 81, 202 lip, 1*5, 320 liquefy, I JO liquor, 171 lisle, 136, 137 literature, 57 little, 209 little girl's room, 199 live, 264. 330 liver, 226 liverwurst, 101 llama, 100 load, 1S4 loaf. 48. 99. 336 loafer, J01 Loam, 336 loan, 61, 10? lobby, 101 loggia, 100 lojic. S2. 368 logician, 368 loiier, 101 lonesome, 131 Ions, 184,36a longed, 262 longer, 368 look. 135,230,344 loose, 61 loot, 102 loquacious, 39 lore, 4S. 295 lose. 112, 114, 224, 261, 321, 330. 337 loss, 114 lot, 357 kmd, 144, IS4, 209, 210 louse, 257 bve, 25, 32, 163. 330 love chUd. 199 loveel, 262 low, 6t, 196, 209 Ion, 141 LSD, 141 lucid, 112 luck, 25, 101,116,195, 207 lucrative, S7

luff, 101 IUK. 147 lull, 14} Lump, 147 hina, 112 lunatic, 87, S8, 137 Lunch. 141 lunk, 147 lurched, 262 lure, Ш lush, 9* ILJSL, 135,173, 174

luxury. 112 lynch, 13S macadam, 195 machiavellian, 136, 138 machine. 42. 255 mackerel, 256 macldnaw, 104 mackintosh, 137 mad, 168 madam. 82, 197 madrigal, 100 mafia. 100 c 137, 176 magician, 55 Magna Carta, 167 magnanimous, 88 magnate, 167 magnificence. 167 magnificently, 191 magnifico, 100 magnifier. 167 magnify. 87 magnitude, 167 magpie, 39 maid, 90 maintain, 136 maize. 104 make, 295. 345 make love. 200-201 ma| adopted. 127 malady, Й2 malaria, 1(10 male orjjan, 199, 200 malefaclion, 87 mallard, 162 malt, ЗОВ mammoth. [01 man, 34. 114. 135, 219, 238, 257. 336, J5V man-eaLing. 123 maneuver, 112. 3J7 mangle, 101 mango, 102. 116

manna, 102 гплппег. 82, 112, 176 mansion, S2 mansuelude, 92 manual, 112 manure, 112, 173 manuscript, ] 12 many, 329 marijuana, 101 market, SI maroon, 161 marriage, Ш marriage of convenience, 116 marshal, 135 manhall, 198 mart, 101 marvelously, 191 mas, 140 mask, 364 masochism. 137 mason, 90 masque, 1M master, 135, 140, 167, 197 matador, 100 matth, 295 mathematics, 140 matter, L99, 200 mattress, 101 maudLin. 138 maverick, 137 may, 67, 115, 567, 26Я me, 108, 234, 247 mead, 46. 320, 337 meager, 364 теакге, 36* meal, 48, 345 mean, 197,294 meander, 13S meaning. 25, 1Я1 meant, 294 measure. 112, 131 meal, 171, 196. 319, 337, 339, 345 mediaLe.. 92 medicine, 81, 82, KB meek. 61 meerschaum, 1O[ meet, 262 melody, 82 meLt. 261. 337 memorial park, !99, 201 men, 257 menstruate, 112 mental retardation. 200 mentians, 221

Index of Modern English Words mentor, 137 menu, 1 Irt mercantile, H)U mercer, 90 merchant., 90 merciful, 132 mercy, 31 merit, 19Я, 207 merriment, 132 mesmerize, 13B mess, 176, 177,205 me i, 2ft2 metal, 82 nietatheory, 127 meteor, 90 metric, 112 me^aninc, JO0 mice, 257 microbLLS, ]27 miJ.century, 127 might. 267 mightily, 191 mild, 2(гУ, J43 mile, 50 militarism, 130 milk, IBS milkman, 124 mi Her, 90 millinery, 137 million, 100 imrnic, Ш minaret. 101 mince, 82 mind, 25, 17ft, 177 mine. 234, 247 mini-skirl, 12$ minister, 198 minor, 87 mint, 50, 51, 137 e, 81 n, 12Б mischief. 188, 139 miss, 135 missal, К 7 mission, 303 mission control, ] 2 l mister, 1*7 mistress, 14[)s 197 moan, 135 mob. 140, 207 mobile.
rtiodestly, 1^2 molded, 13 J mnlcst, 173 mom, 349 Monday, 112 люпеу, D 6 , 137 monger, 50 mongoose, 102, 116 monkey, 46 monsoon. 101 monstrously, 191 month, M2, 344 monthly difficulties, 199 moody, ]89 mutvn, 4 l 4 9Q; 112, 344 nioonliphr, 134 moonshejl, E9, 121 moose. 104, 256 morale, 195 more, 2K9 morynic, 141 morriscm, 136 mortal person. 174 mosque, 101, 116 mosquito, 100 mosquito hawfc, Ш7 mold, 140. 141, 148, 195 motii. 349 mother, 34,41, 44, 46, 160, 344 motion, 133. 135 motion pictures. ] 14 motor. 141 motorcade. T2
t, 104 muslin, 100 mussel, 180 must, 23, 67, 267, 26E 344 mustathio, 100 mustang, 101 mutton, ffZ, 90 my, 234, 247 myna, 102 myself. 3S mystery, J37, 176, 177 nabob, 102 tiaj, 39, 61, (52, 197.207, 329 name, 337, 345, 354 лаге, 140 narcotics agent, 140 nark. 140 narrow, 183, 209 narrowing. 204 NASA, 142 nation, 124 nationalities, 124 nationality, 124 nn i ion hood, 130 nations, i 24 NATO, 141 naiighlY. I8E, 190 navy, 82. 161 nazi. 141 NBC, 141 necessary, 87 need, 112 needle. (12 negro. 100. 107 neither, 366 nemesis. 137 ncn-clsLssica], 128 nephew, .34, 320 nerve, 112 nervous, 87 net, 33. 364 nett, 364 neurotic, 112 ncwl, 295 ne\i, 350 nice. £2 nickel, 101, Ш nickname. 295 niece. 34 nifty, 99 nijger. 100, 107 ni-jh, 340 nipht. 44. 45. 366 nitktrniire, И 7

398

Index of Modern English Words

nines, 205 nitpick, 180, 18L-IS2 nitty. 180 nllly-fiiitly, 180 nix, 101 no, 27 no ore. 279 nobody. 279 nod, 110, 349 noise. 82 none, ) 11 поп -payment, 128 non-standard, 98 noodle, 101 nor, 98 Norman, 50 Normandy. 59 Norseman, 50 north, 4£ nose, 206 nol, 366 nolary, 87 note, 344 nothing. 359, 360 noun, 26 now, 248 NOW, 142 nowhere, 279 nuclear. 140 number, 82 nuncio, 100 n uncle, 295 nunnery, Bl nurture, 135 nylon, 120 nylons, 187 oak, 46, 557 oar. 336 oaths, 360 obdurate, 317 obedience, 81 obey, 323 obligatory, 131 obsess, 11 obstruct. 133 obstruction, 133 obtain, 113 obtuse, 146 oocull, 167 ocean, 46. 82 OLfilOt, !04 ochre, 161 octajius, 110 odd, 61 oddity, 132 s. 310

odor, 173, 174 „dyssey, 136, J 37 of, 28, M5h 163 off, 44 offer, 331 often, 317, 362 ohm, 101, 137 oil, 176 oil painting, 133 OK, 143-142 ohra, 109 old, 27, 115,219,336 old maid. 197 nldtype, 131 olive. 4&. 161 omnibus, HO on, 28, 44, 219 onanism, 137 once, 258 one, 112 open, 255, 337 oj>era. 100. 257 opinion, 82 opossum, 104 opportunity, 113 or, 9R orange, 160,161.295 orangeade, !2S> orchestra, 134 orchestrate, 129, 134 ordeal, 176 ore, 46 organ iKition. 136 orgy. 173 original, 131 ornate, 87 ostrich, 162 otter, 46 ouch. 143 ought, 28, 267, 302, 309 ounce, 112 our, 2U, 247 ourselves, 38 outhouse, 139 oullay, 12B output, 13S outrageously, 191 oven, 90 over. 28, 44, 121, 337 overdrive, 122 over-flawing, 122 overpayment, 128 overthrown, 122 owe, 340 owl, 46 own, 329 own up, 114

ОТ, 46, 257 oxen, 257 oxygen, П9 pace, 295 pacification, 82, 29B pack, 3*9 pad, 348 paddle, 110 pagan, 185 page, Я2 pail, 107 pain, B-2, 131, 168,226 paint. 221 painted, 223 painting, 82 pair, 82 pajamas, 102, 364 palace, Я2 pale, 209 paleface. 123 palm, 3J8.362 psilm tree. 46 palming, 95 palsy, 112 pan, 51 Pan-American. 128 pander. I3B psmtr, 110 panic, 137 pans, 90 pantaloons. 140 panlher, 102 pamies, 199 punts. 136. 137 panty, 202 paper, 82, 187 parade, 100,219 paradise, 102 paraffin, 101, 116 ригаНеIngram, 88 parameter, 112 para-mil Itary, 128 pard, 102 parent, 34 parents. 160 pariah, 102, 197 parishioner, SI parliament, 82 parliamentarian, 129 part, 82 pass, 82, 295 n*ss on, 199 passage, 129 passed, 361 paslal, 100 pat, 143, 309

Index of Modern English Words p*ich, 317 palernally, 132 path, Э4&, 349. 35Б patient, SS, 89 patio, 101 patriatth, Я1 patriot. I1 2 paw, 110 pawn, 110 pay, 81,82 payee, 129 pea. 51 pence, 25, 32. 31, 82 peaceful, 132 peach, 82. 205 pea-jacket, [01 pear, 321, 346 pearl. 82, 226 pears, 311 peasant, Я2 peculiar, 173 pedal. I l l ) , 111 pedestal, 100. 110, 111 pedestrian, MO, I I I pedigree, l i f t . 111, Ш pedlar. 110, 133 pee-pee, 202 peepers, 206 peers, 323 peevish, 133 pelt, 110 perf Я2, ! 75, 310 penalty, 135 penetrate, BS penguin, 55. 116 penis, 200 penitentiary, 140 peon, 100, 110. I l l people. 82. 219 pep, 310 pepper, 51, 102 perceive, E9 perch, 33,256 i»erfeetly, 1*1 perhaps, 19 periwig- 140 perpetrate, 112 person, 114 perspire. 200 pew, 100, l i f t pest, 168 pet. 309 pctly. 168 petunia. 104 pew. 111 phenomena, -57 phew, 143

philanthropist, 1*3 phlHsttne, 137 phony, 205 phuocy, 143 photoRraph, 140 physician, 82, 90 piano, 100. 219 piano 1ед5, 106 piano limbs, 106 piazza, 100 piccolo, 100 pick, 110 pickle, I f t l , 116 pickpocket, 120, 123, 139 picture, S7, 175, 176,351 pie, 110 pie-a-la-mode. 116 piece, 82, 197 pierced, 311 pig, 46, 1Я5, 239 pigeon, 162 pigiet, 130 pile, 195

pilfer, 114

pilgrim, 81 pill, 173-174 pillow, 51 pilot, 110. I l l pimple, 110 pin. 51, 135, 316 pinchbeck, 136 pinchn.it. 134 pine, 162, ЭТ6, 321 pink., 110, 161 pipe, 51 pipsqueak, 99 piquant, 209, 211 piss, 99, 136, 200 pistol, 100, 137 pit, 309 pilch, 50 pilhy, 39 placard, 101 place, 32, 136 plague. 82 plaid, 55, 317 planet, 90 plant, 51, 162,176 plate, 82 Platonic. 136. 138 platter, S2, 90 play, 24, 168 playact, 134 plaza, 100 plea, 135 pleasant, 367 please. Ai, 323, 367

399

pleasure, 341, 351 pliers. 112 plimsoli, 136 plot, 110 plough, 135, 364 plow. 340, 364 plowed, 311 plum, 5t, 161 plumb. 135 plummet, 332 plump, 110, 147, 199 plunder, 101 plunk, 147 ply. 112 podiatrist, 110 podium, 110, 111 ports, 310 podunk, 104, П 6 pogrom, 101 poignant, 210. 211 point, 112 pointed, 39 poison, 131 poisonous, 131 poker, 101 pole. 257 police, 255 policejnan, 120 polka. 101 poll. 101 polo. 102 poltergeist, 101 poly dialectal, 129, 298 pompano. 101 poncho, 101 pooch, 32B poodle, 101 pooh-pooh, 143 pool, 144,311, 326 poo-poo, 202 poor, 81,82 pop, [20. 143 pope, 112 porcelain, 100 porch, 82 pork 82, 90 portfolio, 112 portico, 100 positive, 113 possibly, 19 post, 82 posl office. 120 postgraduate, 12* pot, 309 pot cheese. 107 potato, 104 pot-pie, 120

400

Index of Modern English Words

pots, 90 pouch, 328 poulterer, 90 poultry, 90 pound, 50 роигч, 219 powder room, 106 powerful, 192 powwow, ИМ practice, 364 practise, 364 praise, 136, 157, 173, 198, 295 prattle, 38 pray, 13* prayer, 82. 188 precious, 88 predetermine, 128 predicament, 170, 173, 207 pre-emption, 133 prefers, 221 prelate, 81 pre-owned, 199 presidency, 129 president, 112, 121 pressure, 131 presume, 136 pretence, 364 prelense, 364 pretty, 99. IS9, 191 p r t t a i , 101 prevaricate, 115 prevent, 87 priest, 238 prieMesb. 239 primn donna. 100 primp, ] ]Q prince, gl, 82, 348 prirtLefinj!, 130 principal, 193, 322 principle, 98, 193, 12? prim, 135 piinlatorntm, 131 private. 317 private soldier, 140, J74 prize, 295 pro, 120 proceed, 82 proclaim. 131 pro-corn muniit, 12Я. pnocraslinate, 1J4 prod, lift produce, 367 pnotfuclion. 187 productive, 367 productivity, 367

professional, 120. 140 professor, 140 profile, 100 profound, 367 profundity, 367 program. 195 progress, 109 project, £7 prolix, 39, Я ft prolongation, 82 prum, 206 promiscuity, 163 promote, 87 promptness, 132 prong, 110 proofread. 134 properties, 113 prophet, 81. 88 proposition, 82, 173,20й prose, Я 2 prosecute. 87 prosody, S7 prostrate. 112 protect, 255 protein, 101 proto-Gcrmanic, 128 protrude, 112 protuberances. 298 proud, 81, 8S, 89 prove, SI prowler, 206 pscu do-intellectual, 12$ psyched, 311 psych u l o ^ . 140, 317 pub, 140. 302 public drinking house, 140 publican. 8S puce, 161 pueblo, 101.333,352 pug, 147 pugilist, 112 pugnacious, f 12 puke, 110.203 pull. 135, 144, 310. 326 pulpit, 87 puma, 104 pumpernickel, 101 pun, ПО ptinch, [02 punctual, I 12 pundit. IdZ. 116 punuem, 112, 209, 211 ptmi.Hh. 113, 11J, 367 punishment, П4 punishments, [13 punitive. 367

punk, 110, J47 punster, 130 punt, 112 purchase, tlipurgatory, 81 purloin, 114 pus, 200 push, 82, 208 PUSH, 142 pushed, 262 puss, 114 pussy, 110. 201 put. 135, 309, 310, 311 put-il-off-itis, 130 put to skep, 199 puts, 310 putt, 309 puttee, 102 puzzle, 110 puzzled. 114 pygmy, 112 pyjamas, 364 pyragencsis, 132, 138 quack, MK quad, 308, 348 quadrangle, 140 quadrille, 101 qualm. 348 quarrel. 3f>8 quart, 343 quarter of, 107 quarter till, 107 quarltr to, 107 quartet, 100 CtuarLi, 101

quean, 196 queen. 198 queer. 110 quiet, 146, 190,321 quickie, 130 quickly, 20fi quiet, 87, 185, 209 : 211 quinine, 104 quirk. 110 quirt, 101

quisling, 137 quit. 82, 109,302,309, 317 quife. 27 quixotic, 136, 138 110 rabbi, 102 rabbit, *6 rabelasian, 13.8 race, 61, TOO

Index df Modern English Words racial, 132 racial slock, 131 rack, 101 racoon. 104 radar, 14 J, 195 ragp, 114 rah, 143 railroading, 298

raise, 61. 109, 115.321

raison d'etre, 116 rake. 207. 345 ramble, 110 ran, 355 rape. 112 rapid, 115 rip idly, 208 rapport, 112 rapresentative, 141 rapt. 112 rapture. IJ2 rascal, 197 rit, 207 rather, 27 ration. 116 rattan. 102 raven, 112 ravine, 112 ravishment, 82 read. 221, 2*1,319, 345 reads, 219. 224 real. S2, 100 rc:rt|y, 192 reap, 2fi|, 354 rear, 295, 321 reason, SI rebec. 101 rebuke, [14 receipt. 317 receive, &9 reclaseiry, 123 теоа^пЬл, 89

reconcile, 223 recreation, 82 red, 124, !60 redbreast. 123 redcoat, 123, 139 reddish. 124, 130 redeem, I3(> red е1пл ificat ionabili lyF 132 red worm, 107 reef, 61 reek. 173. 261 recked. 2(M refere*. [40 reflect. 25

regatta, 100 register. 82 rehearse, 136 reindeer, 61 reinforce. 112. П4 neinforce merit, 113, ] 14 reject, &7 re joke, 136 relative, 34 relieve, 112 religion. Si. £2 remain, [36 remedial, 367 remedy, J<57 remember, 82 remind, 177 reminiscence, 133 removal, II3 7 114 remove, 114 rendezvous, ISS rent, Я1 repair, 1 12. 1 15 repeat. 367 repetition, 367 repetitive, 367 replacement, 124 report, 1Й5, 187 reprehend, 87 representation, 82, 167 representative, 28 reprimand, 114 reprove, 114 repEidinle. 110 require, 112 rescue, 1 iti resemble, 25 resentful, 132 reset. l?3 reside. 112 resign. 317, 368 resignation, 36$ resilient, 185 resist, 150 resistance, 150 resolute. В Я resolve, 317 respite. 82 responsefut, 1?2 responsible for, 157 restrain, 136 result. 185, 206 resume. 223 resumed, 122 resurrection. Sift, Ш relait, 135 letaliale. 170, 173 reticent, 39

retreat, 19* retrorocket, 42 retro-rorkel, 128 revealed, 323 revemic, B2 reverence, 136 reversal, 125 reversals, 125 reverse. 125 reverses, 125 revise, 150 revision, 150 revisionism, 298 rcvok. ПЗ revolution, 1J3 reward, 170 ribs, 185 rice, 46 rich,, 81 rid. 61 rtddle, 135 ride, 177, 260: 261 rille-ranse, 120 rift. 61 rig, 110 ri£hL, 192, 316,368 righteous, 368 rins, 135, 261 rip, 1 [0 rip off, 114, 185 rip-off, 163, IS5 ripped. 2b2 rise, 135, 261.295 risen, 261 risk, inn river, %2 rivet, 135 roach, 323 road, 309 roam. 1 Ю roar, 143 roast, 82 rob. 81, 249, Ш robber, 28!> robbery, 114 robe, 82. 135. 344 robin, 49 robrtl, 101 roc. 102 rock, 25. 34, 55, 20S. 226. 239, 349 rock резк, 55 rocked, 223 rocket, !00 rodeo, 101 roll, 223 rolled. 223

401

402

Indc* of Modern English Words

romance, 82 rood, 344 rook 344 man, [35 root, 99 rope. 744 rosary, 87 rose, 81, 161,261.337 rotate, I15 rote, 110 rotten, 61, 312 rouble, IftI s >2, 351 ush, 1S5, 209,210 ground, [38 ronghgrind, 123, 13* rough-ground, )33 round-windowed, 126 routed, 262 rove... 110 row, ПО. 261 myal, Я 2 royally, 132 nibbed, 262 rubber, 331 rubbers, LB7 ruby, S2 rucksack.. 101 rude, 304 rue. 261 ruffle. 110 rug, 147 rumble, 147 rumbullion, MO nimp, 147 mn. 25, 135,261,355 rundown, 136 running. 131 runs, 3i0 rush, 136 russet, 161 rust, 110, 135 Sabbath, 102 saber, 101 sack, 114.337 sacrament, £1 sacrifice, 135. 137 sad, 1 15, 302 sadism, 137 safari, 101 safe, 82 safeguard, 135 saffron, 101 sag, 34» said, 329 sail, 48, 175, 176, 195

sailboat, Ш saint, Я1 sake. 102 salad, 82 salamander, 102 salmon, 46.. 161, 165,256 salt, 4ft, 131,336 Sam hill, 202 same, 61,345.346 samovar, 101, 116 sanctimonious, 173 sanctuary, 111 sand, 348 sandal, 102 sandblind, 128, 192 sandwich, 137 sang. 37, 219. 222,260 sanitary engineer, !99 sanitary napkin, 199 sapphire, 102 sari : 102 sarong, 102 sarsaparilla. 100 sash. 101 sassafras, 100 Satan, [02 Kate. 326 satellite, 195 satin. 32 saucer, 82, 90 sauerkraut, lOf saulerne, 137 savoir faire, 116 saw, 223 saws, 224 say, 264, 323 scab, 61 scads. 110 scale, 328 scales, 61 scare, 61, 62 scarlet, [02, 161 xuatier. i 10 scavenger. 133 scene. 317 scent, 176, 177 schedule. 32S scheme, 92, 173 stherzo, 100 schnozzola, 206 schooner, 101 scientific. «8. 92 scientist, 132 sclerotic. 333, 35? scoff, ЗЛ9 stold, 39, 196 scoop, 1 Ю

scorch, 110 store, 61 scoundrel. 110 scowl, r i ! , 110 scrap, 61 scrape, 166. 261 scratched, 310 screech, 42 screwed up, 114 script, 87, 309 scripture, B7 scrutiny, В7 sculptor. 133 sculpture, 82 sea, 46, 48, 323, 339,366 seal, 48 season, 82 seal. .«.61, 199,319, 326, 345 second, В2 secrete, 135 seculaj, 87 sedan. I )0 sedation, 112 seduce, 173 see, 12!, 261,366 seed, 46, 261 seed bearing, 122 seem, 230 seemly. 61 seems, 323 seersucker, 102 see-saw, 107 seethe, 261,321 sell, 37, 113,224 >;eme.4ler, 101 .Hemi-cij'cle, 12S seminar, 101 senior citizen, 199 sense, 348 sensible, 166, 207 sentence, 114 sententious, 39 sentimentally. !30 separated, 222 sepulchre, в I serape, 101 seraphim, Ifl2 serenade, 100 sergeant, 140 sermon, 81, E2, £9 servant, 54, 90 sesquipedalian, 110 sestet, 100 set, 26?, 326 setback, 123, 136 settee, 33

of Modern English Words settle, ] 11 setup, 138 seven, 45 sew, 46 sexual relations, 199 shabby, ПО, 166,207 ebacfc, 110 shackle, Э2В shade, 323, 346 shaft, 32Я shall, 195 shake, 135,261,329 shaken, 261 Shakespearian^, 130 sbafc, 101 shut], 2Я, 2?, 267, 268 shallow, 185,209 sham, 95, 110, 207 shame, 302 shamefaced, 126, 192 shampoo, 102 shamrock, 55 shanghai, 138 shape, 34, 323, 328 shaft. 135 sharp. 44, 114, 146, IS5, 19*. 209 shave, 261. 32S shawl, 42, [02 she, 234,238,247
shrift, 328 shrill, 209,210, 211 shrimpboal, 124 shrimpburger, 129 shrink, 261. 309 shroud, 171,257 shruj>, I JO shucks, 201 shuffle. 207 shuffling, 95 shut, 135 shuttle, 135 shy, 32 8 shy lock, 136 sick, 109, 125,219,302 sick to the Stomach. ]07, 203 sicken, 125 sickening. 125 sicking, 125 sickingen, 125 sickly, 130 sideways, 25S siesta, 100 sieve, 317 sieves. 310 sight. 366 si^htsee. 134 sign, 317. 367 significant, 367 silence, HI silhouette, 137 siJk, 102 silks, 187 Silly, 146, 17(1 silc, 101 siiver, 4S, 50 simple, It2, 146 simply. 192 simultaneous. 131 sin, 309, 326 since. JJft sir fnl, 13(1 sing, 37, 44, 114, 134, 260, 261, 341 singer-, 134 singin'. 266. 360 singing, 265, 266h 360 single, 82 sink. 261 sip, 303. 304 sir. 82, 311 sinocco. lot sister. 54, 46. 61, 160 sit. 112, 261, 337 six-wheeled (ruck. !Z5 size. 112

403

c, 143 skate, 101 sketch, 101 skid, 110,328 skidoo, 206-2U7 skill, 61, 328 skiJlct, 107 skim, 32E skin, 61, 32S, 337 skinner, 90 skinny, 114 skip, 328 skirt, 61 skit, 32S skulk, 135 skull, 110 skunk, 104, 116 sky, 61. 328 slab, ПО slack, 295 slain, 329 slake, 295 х]апК. 98, 110, 204 slattern, 197 slaughter, 61 slave, 82, П6, 137 slay, 261 sled. 101 sleep, 41, 44. 259, Э20 sleep with. 199, 200-2У1 sleepwalk, 134 sleet, 135 sleighs, 311 slender, f 14 slept, 259 slid. 2S>4 slide, 261, 294 slim, 114, 166, 207 slog, 110 slo3an, 55, 116, 177 slo^ip, 101 slouch. IIO slow, 146, 156 slum, [10 slumber robe, 199,201 slur, 101 slush. UO slut, 110. 197 sly, 61, lift, 190 smack, 143 small, 146,209,210 smart, 114, 146,210, 2Vi sroaze, 140, 141 sniearcssc, 107 smile, 175 smirk, 173, 175 smile, 261

404

Index of Modern English Words

5ТПОБ, 120, 140, 143 smoke, 120, 140, 141, 337. 344 smooth, 185, 209 smus, UO, 173 smuggle, 101 snafu, 141 SNAFU. 202 snail, 34 snake. 46, 185 snake doctor, 107 make feeder, 107 snap, 101 snap beans, 107 snapped. 262 snare. 61 snarl, 147 .snakh, 101, ПО, 114 sneak, 110 sneer, 147 snicker. 143, 147 snide, 110 sniff, 147 sniffle, 147 snigger. 147 snob, 110. 147 snoop, ][!1, 147 snore, 143, 147 snori, 101, 141, Ш , 147 snout, 147 snow, 44.45.46, 10!. 135 snowman, 124 snowshoc, 124 snub, 61. l i f t , 147 snuff, 101 snuflle, 147 snuggle, 147 so, 9S. 34! so lonjj, 27 soap, 135 sob. 143 SOB, 202 social disease, 199 social diseases. 200 sock, 51 sodden. 321 sodomy, 136, 137 sofa, 101 soft, 156. IBS. 20S, 209, 210, 211. 349 sojourn. 82

wJa«. 82

solar, 87 solar system. 90 sold, 37 solder, 135 soldier, 82, 351

lid. Я2 solitary, 87 solon, 137 solve, 112, 185 sombrero, 100 somebody, 27? someone, 279 somersault, 185 sometimes, 25 В somewhat. 27 somewhere. 279 son, J4, 46, 160, 366 son of n sun, 202 sonata, 100 .Hung, 349 songfest, 130 songs, 219 son-in-law. 46 sonnet, |00 soon, 344 soot, 344 sop, 349 soprano. 100 sorely, 192 sorrow, 329 sot, 81 sound, 48. 82 sour. 196,209 sour milk cheese, 107 south, 143 souIhwards, 25Й soviet, 101 sow, 46. 162,261 space age, 109 spaced <j"l, 166, 207 spacious, Я7 spade, 100 span, 261 spark, 135 sparkle, 130,222 sparkplug, Ш spartan, 13fi, I3S, I4S spat, 147 spatter, 147 speak, 261.323, 345 spear, 55. 33V special student. 199 sped. 2ft2 speech, 344 speed, 339 speeds. 311 spend, ?0Y spew, 147, 261 sphere, 333. 352 spider. 107 spin, 261 spinach, Ю2

spindle, 333, 34L spinet, 100 spinner, 332 spire, 135 spirit. 25, 147, 174, 185 spiritless. 88, S9 spiritually, 132 spit, 120, 147,224 .spitz, 101 splash, 293 splashdown, 89 splint, 101 split, 1*5 Kph[rs;e, 143 Sponge, 81, 135 spook, 101 spool, 101 spoon, 90, 344 sport, 112 sportsdom. 129 sportsman, E Ы spol, 121 spotlight- 121, 138 spout, 147 sprain, 110 .spray, 147 spread, 222, 293 spree, 110 sprig, 141 spring, 46, 261 sprinl, 61 spry, 1 Ю spume, 147 spurn, 147, 261 spurt, 147 spulnik, 101 spulter, 147 squabble, 166 squadron, 100 squalid, 348 squander. 110 square, 82, 185, 219 squash. 104 squashed, JI0 squat, 34R squeak, 143 squ&ezed, 311 squelch, 143 suuclcrred, 309 squid, 110 squire. 82 squirrels, 312 squirlers, 312 stack, 61, 309 Madia, 257 staff. 337 stage, 112

Index of Modern English Words c, 134 staid. ] 12 stamina, 257 stampede, 101 stand, 25, 112, 261 Standard, 1I2, 225 standby, 121 stanza, 100, 112 slar, 90, 219 stare, [35 stark, 166 starling. 48 start, 176. 177,308 started, 262 5<»rvc, 173. 175.215, 222, 223-224, 261 starved, 222, 224 stale, 82, 112, 157, 323 statement, 114, Ш static, 112 station break., 121 statistic, 112 StulUC, 112

steadfast, 126, 130 steak. 6 1 , 319, 346 steal, 114, 135, 163, JS5, 261, 337, 345 steam bath, 124 steamboat, 124 steamfittcr, 124 steel. 46 steeled, 323 sleckshafled club, 125 sleep, 209 steer, 4* stench, 170, 174, 295 stentorian, 138 step. 261, 337 steppe, 101 stetson, 137 stevedore, 101 stew. &2 stick, 205 still. 205 stimuli, 257 sting, 261 stingy. 99 Mint, 135, I73.2Sl.295 slinkcr, 99 stint. 2*5 stipple. 101 stir, ] 3 5

stirrup, 329 sttlch, 295 Stoker, 101, 133 SlolcL 26 I

stomach, Hi sLunc, 135,226, 344 stool, 53, 171, 174, 344 stop, 177, 349, 357 stoney, 361 story, 55, 8 1 , 92, 364 stouthearted, 125 stove, 90 straight, 185, 205,219 strain, 136 strand, 48 strange, 82 stream, 135 strident, 209 strike, 261 string beans, iO7 stripped, 309 slrolj, ПО strong, 310, 368 slrongarm, 123 stronger, 36Й struck. 16$ struggle, 110 strung out, 17Й sLutco, 100 studio, 100 study, 82, 224, 225 stump. 147 stunt, 147, 295 stupendously, 192 stupid, 114, 146 stupldenls, H i slupor, 87 subdivide, 87 *uhj[]£a.ie, 87 submarine, 140 submit, 87, 114 subordinate, 87 subscribe, 87 suhslance. 34. 82 substitute, 87 subsurface, 12-8 suburbanite, 130 succeed. 224 success, 173 successful. 132 sued net, 39 succotash, 104 succour, 112 such, 359 suck. 135 rudder, B2 suds, 110 sugar, 92 5UEI?e5tive. 173 4uicf<Jal, 129 sulk, 11()

405

sumac, 101 summary, 87 summer, 46 summon, 136 sump, 147 sumptuous, 82 sun. 41, 44, 90, 337, 366 sung, 260 sunk, 147 superabundance, Я7 supercilious, 167 super-heated, 128 superman, 116 suppeditatc, 92 supper, B2 supplicate, 87, 112 supplies, 187 support, N 2 . 13ft suppose. 259 suppress, B7 supra-national, 128 sure. В2 surgeon, 82 surly. 153. 176, 190, 197 surrender, 114 susceptible, J33 svelte, 333. 352 swag, 34« swagger, 348 SWilrt, 61 swallow, 4Я, 261 t. 34Я swamp. 34K swan, 33ft, 348 swear, 261 sweet, 209. 210 sweet music, 185 swell. 261 swim, 2ftl swindler, 133 swine, 90, 256 swing, 261,303, 309, 33? swinged, 261 .swinyer, 199, 201 swinging, 163 swivei, 129 swizzle, 110 swond play, 168 syndicate, !73 syrup. 42, 101 systaltic, 112 system, 92, 112 tab, 114 table, 82. 174 tabled, 139

406

Index of Modern English Words

labno. 42, 102, ]37, 14Я,

195 taciturn. У) tackle, m i

taffeta, 102 tail, 135, 197 tailor, 90 take, 61. 323. 345 talc, 345 talk, 38, 135. ЗОВ, 317, 34S talkathon. 129 talkalive, 39 tall. 34S tame, Mfi tan, 161 tanjsy, 209 tantalize. 13 В tariff, 109 tarry. 110 tart, 196,209.210 lasV, П taste, 82 tattered, 61 taunl, 110 tawdry, 139 tax, 32, 136 taxi, 120.207 taxi-mete г cabriolet. L20 ТВ, 202 tea, 102 teach. 29}, 321 teaches, 224 tear, 261 tearful, 146 tears. 32? tcchec, 143 teeny. 42. 120.293 Leeter-board, 107 teeler-totter, 107 teelh, 203, 206. 257 teelhe, 295 tclcprionjejtlly, 29 Я 1rl с vision, 42, Ш tell, 115,337 temperate, E7 ternj>o, 100 temporal. S7 tempt. 81 ten, 45. 302. 303, 320 tennis, 2fi tenuous, 114 termagant, 196 terminal case, 199 terminate, 1t5 terribly, 192 terrific, 207

terrifically, 192 terse, 39 testicles, 206 testify, 87 tctc-si-lcfc, t I fi thane, }29 thank, 171 thanks, 309 that, 244, 24 8, 290, 309, 331

thalch, 46 the, 115,219.226,240,

Ml. 244.290. 331 theater. 365 Iheatre, 365 1hee, 247, 248, 249 theft, 114 their, 61.98. 234,247 theirs. 234, 247 them, й I . 234, 247 ihen, 24S, 331 there, 24S, 331 thereabouts. 131 th&se, 244, 248, 290, 331 they. 51, 234, 247, 359 thick. 181209,309, 359 thicket, 130 thimbie. 112, 333, З Л thin, 114, 185,209,302 thin board, 55 thine, 247 ihing, 112.302,321,341 think, 115.259 third. 341 32-fceyed typewriter, 125 this, 244, 24Я. 290, 302, 310,331 those, 244, 24S, 290,331 thou. 247, 248. 249 though, 61, 115, 311, 331, 365 thought, 226, 259 thoughtful, 131 threads. 1 [4 thrice, 258 thrift, Ы throb, 143 throne. 31H-319 thronipi, 310 through, 28, 302,311 throughout, 122 through way, 125 throw, 321 throw up. 123, 139 thrum, 143 thrump, 143 thrust. 61

thug. 102, 137 thumb, 112, 317 thunder. 320 thimk, 143, 147 thus, 248,331, 337 thwack, 14Я thwacked. 310 thy. 247 thyme, 317 tic. 110 tickle, 110 tidal, 132 tiddly wink. 110 tide, 337 • !••!•:•

6 1

tiff, 110, 166 tiger, J<5, 102 light. 61, 99 till. 61 timber, 46 tin, 48, 50 tincture, 87 tinkle, 293 tiny, 42, 166 lip, 110, 309 tipsy, 199 tirade. 100 tire, 364 tired, 219 titanie, 13fi title, S2, 367 tiLttlar, 367 to, 75, 115. 163,366 toad. 110 toast, «2. 110 Lobaccu, 46, 104 toboggan, 104 to-do, W toilet, 186-187,188 toll. 50 Torn, 219 tomato, 104 tomb. 112 tomtom. 102 tone. 131. 135 tongue, 185 too, 3fifi tooth, 45. 257, 295, 320, 321. 344 top, 325, 348

Torah. 102

larnadn, 100 tortilla, 100 tortoise, 46 Lass. 349 tote. 42, 109 totem, 104

Index of Modern English Words touch. л28 touchy, J66. 207 tough, 302 tournament, 81 toward, 12Я (fitter, 81, 82 town, 321 tract, 87 tragedy. 82 train, 302 (ram, 136, 137 trans-Allan lie, 128 transfer, 204 transition, 133 translate, 1S5 transmission, 133 transmit, Ш transportation, 298 trapeze, 110 trapezoid, 110 trash, M0 IraVcl, Ы traveled, 364 travelled, 364 travertine, 100 tread, 261 treason. 135 treasure, 135 tree. 162 trees, 323 trek, 101 tremendous, 207 tremendously, 192 trenchant, 39 trial, ISSU

lrin.1 balloon, 116 tributary, 87 tribute, 50 trick, 114, 135 trim, 1 Г0 trio, 100 lrip L 168, Ш tripod. 110 (ripped. 224 troiia. 101 trollop. 197 from bone. IIX) trot, 1*7 troubled. I [4 trout, 256 trtid™, ПО Irue, 321 turtle, П 2 irull. 197 trunk, 147 trusl, 61 tryst, IBS

tsk tsk, 143 tuberculosis, 112 Tuesday, 112 tug, 147 tulip, 102 (umble, 147 lum e went, 112 tummy, 202 tuna. 256 tundra, 101 tunny, 100 turban, 102 turk, J37 turn down, 206 turn off, 20fi turn on. 1S5, 207 turnpike, 140 turns, 224 tush, (43 twaddle, 110 twain, 329 twal, 201 (week. 110 twite, 258 twig, 141 twinkled, 219 twirl, 140, 141 (wish, 143 twist. 141 twitler. 1
undiscernina, 88, 89 unequal, 128 unfeminine, 197 ungrummatica], 204 un happiness, 24 unhunh, 143 "uitcJIiviar, 128 unicom. 81 unintelligent, 114 union. 112 universe, 90, i 12 university, 140 unladylike, 197 unthinkable, 132 until, 316 unusualncss, 132 up, 139 upchuck. 199, 203 uphold. 12-8 urinate, -^00 urination, 200 us, 234 use. 82 usherette, 130 usual, S2 Utopian. 13B vagina, 200 valiant, 88 valley, 3J8 vamoose, 101 vamp, 197 vampire, Ш2 vandalize, 13*, 148 vane. 323 vanilla, 100 vapid, 20У varlet, 112, 197 vassal, 112 vastly, 192 vat, 323 VD, 199,200,20: veal, 82, 90 vegetable plant. 174 veins, 311 veldt. 101 velocity. 115 vend. 309 VendclM, 100 venison. 82, 90 vent, 323 verb, 26 verbosity, 367 verdict. 82 vcrmillion. 161 vermouth, 101

407

40S

Index of Modern English Words

very, 27, 192 veteran, 140 viands. IM vice, 171 vice-cfoairmatt, 128 victoria, 136 vigilante. 101. 116 villain, 197,207 vintner, 90 viol da gamba, 100 viola, 100 violently, 192 violet, 161 viper. 154 virago, 1% virgin, 81 virtue, 135 vision, 341, 351, 3*2 vista, 100 vivisection, 133 vixen. 323 vodka, 101 voiced, 311 void, L99 voiding, 200 volt, 137 • voiubic, 39 volume, Я2 volunteer, 166, 207 vomit, 203 voodoo, 360 vulgar. 98, 1*7 wad, 110,308,348 wadJk,34S wade, 261 waffle, 101 waggon, 364 wagon, 46, 364 wail, 207 wainscAl, 10 i wait, SI, 82, 1 2 5 . 2 5 7

wailed, 219 wailer, 90. 125, 238 waiter's, 125 waitress, 239 wait's, 125 wake. 261,2*5 walk, 24, 115, 135, 177, 31S, 34S wall, 51 walinowcr, 197 walrus, 101 waltz. 101 wander, 348 wangle, 99

want, 61 ww, &l, 1S8, 308,323, 34B ware, 4*, 50 warm, 156, 1Б5,209, 2)0 warn, 135, 34S warp, 348 warl. 34* was. 219, 264, 265, 267, 268.321,348

wash, 135,224,261

washroom, 199 wasp, 34S walch. 115,295,34» watched, 219 water pistol, 124 waterbed. 124 water-clock, 120 wjitcrdoset, 186 watt, 137 WAVES. 142 wax. 261. 348 way. 329 WC, 202 we, 234, 247 wrat,61,3l9. 323, 345 weariness, 13 L weasel. 46 weather, 135 weave, 46, 261, 295 weave!, 337 web, 295 wee wee, 199 week. 337 weep, 135,261,344 wee-wee, 202 weigh, 25. 261 weiner, 136. 137 well-penciled student, 126 welt&chmeri, 1 16 wench, 196 wend, 264 were, 264. 321 west. 4S wet, 309 whale, 48 wharf, 34R what, 2S3

whrjt, 46, 4S wheel, 46 when, 237, 2ЯЗ where, 237, 2B3 whet Ktone, 122 which. 98 while, 115 whip, 154 whirl. 141

whiskey, 55 whisper, 293, 358 whispered, 222 whistle, 143 while, Ш white meat, 106 whizzed, 310 who, 12, 283 whole, 318, 335,336 whom, 12 whore, 135, 197 whose, 96 why, 237, 283 wide, 209 widow, 46 width, 166, 310 wife, 44, 154,325 wild, 185 wildebeest, 101 will, 28, 29, 115,267, 268, 337 willow. 46 wilt, 109 win, 261 wind, 261 windbag, 3? window, 61 wine, 50, 219 wink, 135 winter, 46 winterize, 132 wire, I,i5 wisdom, 129 wise. 44, 146 wisecrack, 99 wise-minded, 122 wish, 135 wisp, 1 (0 witch, 196 with, 2.B, 115, 126 withali.331 withdraw, | | 4 , 126 withdrawing, 114 wiihhold, 126, 123 without, 126 withstand, 126. 128, 331 witness, 81, 135 witty, 146 wizard, 140 wolf, 46 woman, 35, 162. 197, 239, 341 womb. 336 wombat, MA women's liberation, 140 wonder child. 116 wonderfully, 1LJ2

Index of Modern English Words wondronsly, 192

ivouil, 337 woodchutk, 104 wtujilsman, 90 wool, 302 woolen, 129, 364 wutilcns, 312 woollen, 364 . 135, 336

/, 39

wort. 112. 222, 223-224 a'orld view, I 16 worship, 135 worm, 38, 50 wove, 261 woven, 261 wow, 143, 302 WPA. 148 wreak, 26 i wrestle, 185 write. 4*, 135. 2 60, 261

write in, 13 writhe. 261 легок, 137 yank,110 yap, 206 yawl, ] 0 l ye, 329 yea, 319,. year, 302, years, 323 yell, 261 yellow, 160. 32* yelped, 309 yes, 27, ,108 yet. 98 yield, 261 yippee. 143 yodel, 101 yoga, 112

409

yoke, 4fi, U 2, 320, 329, 344 yon, 329 you, 234, 247, 248-249 your, 234, 247 your*, 234, 247 yourself, 38 jxbra, 142 zeilacist. 116 zen, 102 zenifh, 101 *eppetin, 137 zero, 101, 116 *ij='-*;iS> 10t zinc, 10 J zip, 302 zilher, 101 uoo, 140 garden, 140 r, ] ] 2

SUBJECT INDEX

Ablaut. 260 Г Academic francaise, 94 Atcademia del I a Crusca, 94 Acronyms, 120, 141-42 Adjectives; comparison of, 257 conversion to noun*, 140, 174 weak-strong, 244-45 Adverbs, 258 intensifies, 190-91 Aethelberht, 56 Affixes: derivational, 124-33, 366-68 inflectional, 124-25, 234^9. 259-67 fee also Grammatical change Affricate, 307, J5I African languages: borrowings from, 109 influence on Black English, 107-109, 275, 277, 360-62 Afrikaans, 4S Ain't, origin of, 277 Alfred ihe Great, 56, 59, 63, 3J<J Alliterative Marie Arthurs, 15 Allophones, 303 Alphabet, 312-17 of, 315 of, 314 set.' also Writing Alveolar. 303 Alveo-psJatal. 308 Ambijjuiiy, 30 Amelioration, 19B American colonial settlement, 103, 3?7-5S American dialects, 103-109, 357-61 Dictionary vf the English ge, An, 364 American English: altitudes ifiwanJ, 104 pronunciation of, 357 vocabulary, I OS American Heritage Dictionary of the English Language, The, 99

American Indian, borrowings from, 104 American РЬЕМо^Ы Association, 367 American spelling, 363-66 American SpeUing Book, 364 Anaphoric do. Ill Angles, 5,1-54 Anglo-Saxon Chronicle, 59, 60, 71 Anglo-Saxon Heptarchy, SS-5b Angular syrus, J3-I4 Animal languages: see Bee; Primate Aii'tflles Angto-fiaxunici Breves. 71 Antony and Cieopaira, 135, 249 Ape: see Primate Aphesis, Ш Apocope, 140 Arabic, 45, 46 borrowings from, 101 Aramaic 46 Artjot, 203 Aristophanes, 204 Aristotle, 91 An,' of English Poesy, Tin'. 53 Artluiiirctiiti Merlin, 15 Arthur, King, 75 Articles, 236^9 Middle English, 245-49 Modern £m;liih, 239-4J Old English, 236-*J Arlifactual words, L16, 195 At Yon Like It. 249 Ash. John. 166 Atheistjn, 59 Attic Greek, 4* Australian Aborigine, borrowings from, 102 Auxiliary, 267-74 Early Modern English, 268-74 Middle English, 272-74 Modern English, 267-74 negative», 278-SI Old English, 270-72 questions, 2SI—B6 AwdcLey, John, 204 411

41U

Subject Index

Babel. 1 I Hack vowels, 144, J09 Hiickioi illation,

133,223

Bacon. Francis, 3J6 Bailey. Nathan. LC5 Balto-Slavic, 47,49 ВаГОПв Wars, 7 [ Barton, John, 76 Base Rules, 217-2(1, 227, 266, 268-73, 279, 282, 28d-&S Btlllk fff Bnmnanimrg, 60, 352 Bauph, Albert С , &2.253 deletion of, 275-77 forms of, 264 invariant in Black English, 276 rebiion lu ain't, 277 Bede, 54, 56, 60, 7 *, 242, 357 Be«, language of, 15, IB, 142 Beetliovcn, Ludwig van, 23 Behaviorisi psychology. 26, 32-33, 154-56 Bengali, 49 Beowulf. 4,56 Bible, 150 BflftbW, 30S

ВЫ, 360 Black Americans, hislory of, 5, 107-109, 360-61 ВЫ-к Death, 69, SO. 81, в? Black En^ish: African origins, IO7-L09, 360-61 f, 274-77 of, 359-62 see alto Social dialects BleruK. Ш , l40^tl t, Thomas. 16S a, Queen, 53 Boethius, 91 Bate Named tkt GovertSOUT, The, 93 Bticsk of the Dockets, Tin, 297 Borrowing: derivational ;iffi3i«, 125-33 euphemisms, 200 H>n[j\. 253

vocabulary: African languages. 109 American Indian languages, 104. 192, 256-57 Arabic. 101 Australian Aborigine, 102 Celtic, 56-57 Danish, 61 Dravidian LanjjilaeieS, 102

Flemish, 101 French, Й7-ЙЙ. R1-S3, 90,92. 135-36. 149-W. Iftl Frisian, 101

Borrow] ПЕ vOcabutary (cont.) German, High, 101 German languages. Low, 101 Greek, 92 Hebrew, 102 Hindi, 102 Italian, 100 Japanese, 102 Latin. 52, 57,68, 87,92 Malay. 102 Persian. 102 Phmfeuttch, 101 Polynesian languages, 102 Kcim;myr 102 Sanskrit, 102 Slavic languages. 101 Spanish. [00 Tamil, 102 Telegu, 102 Tibcto-Chincsc languages, 102 Turkic, 102 Bruin, anatomy of, 13-15 Breton, 48 BrewerK Guiid. 73 Bridges, Robert, 367 Briganles. 52 British spelling, 364-155 Brni, 74 I!LI]L\L: [LIH.

47

Buliokar.Juhn, 164.165 Bullokar. William. 363 Hiitlcr. Charles, S4, 364 fiicdmrtn, 56, (5(1, 242 Calvin. John, 9! C'ambretisis, Giraldus. 74 Cant 203-204 Canterbury TaUs, The. BS, 249, 297, 29S. 340, 351.352 (Tarneg.Ee, Andrew, 367

Carr, E.H..S Carr.CT,, 121

Case: function of in Modern En.elivh, 234-38. 252-54 function of in Old English, 234-38, 252-54 lass of dative, 237, 255-56 prepositions, 236-37, 252-54 Cassidy, Frederic G,, J06 Causative di>. 276 Caxton, William, 80, Bl, 86, 91, 92,93 Celtic, 47,4? borrowings from, 54-55 Cells, 51.52 Central vowel, 30У

Subject Index CrnLim languages, 47-48 Cerebrum, 5 3—15 Chaucer, UeofTrey, 5, 29, 30, 75, 80, 85, 92, 249, 349, 350 Cheke, Sir John, 8fiT 363 Child language, 37, 355 Chinese, 45 borrowingifrom, 102 Chomsky. Nuam, 28 ChristianiIy, influence on English, 56-57 Cicero, 9 ] Classical languages, prestige of, 11, L2,

20,24*25,87-89,91-92, 138

Claudius, 51 Clipping, [39-40 Close к oils. 72, 73 Cnute, 62, fi3,66 Cockeram, Henry, 164, 165 Coles, EJishs, 94 Coles, Robert, ]79 Cofjnare words, 22 in Cpcrniiinic languages, 48, 50, 325, 326,328 in fndo-European languages, 44, 45, 46, ,120-321 in reconstruction of culture, 44, 46 in West European languages 48 Color words: devel[]pment of in Modern English, 20H metaphors for, 211 Old English, in, 160 Comparison, history of, 257-58 Compendious Dictionary of jhc English LnnwaxL; A, 166 Compound [tig., 119, 120-24 in Modem English, 121-2? in Old English, 122 stylistic consequences of, 121, 148 versus [earned words, Я8-Я9. 115 Consonanls: Early Modern English, 350-52 Middle English, 341-42 Modern English. 307-309 Old English, 327-32 symbolism of, 142-47 Continuants, 307 Conversion, I3J-38, 174 Copland, 204 Copltc, 46 Cornish. 48 Crashaw. Riuhard, 182 Cratylus., J Crecy, 71 Creole [jLn^tinprs, lus-tf)p Cromwell, Thomas, 93 Cursor Uundi, 75

413

Cu&hitie, Cynewuir, 56 Сгесп,45, 47 s, Simon, 364 Danelaw, .4 V Danish: borrowings from, 61, 68 conflicts with, 58-60 invasion of, 59 Deep slrjcture, 31, 216-2], 236-40, 266, 26Й-8У Definitions, structure of, 33, 155-59 Defoe, Daniei, 95 Dental preterit, 294 Derivation: and backformation, 133-44 history of, 124-33 phonology of, Э66-6Я see и!ш Conversion l>crivalional suffixes, defined, 124 Descriptive grammars, 2, 26-28 and usage. 96-98 Dialects: American, origin of, 103-109, 357-61 British: ,t1liludes toward, <)2-l>i, 104-105 Farly Modern English, 92-94 Middle English. 64, 86, 245-47. 262-65.335, 343-44L 346 Old English. 52-54,64 Micial, ld*-50, 166, 174, 203-207, 247,254,258,261, 274-78, 280-81, 323. 334-35, 345, 347. 348,349,356-57,358,359-61, 362 and style. &7-ЯЗ, 95-99,112-16, 121, 132, 138, 140, 141, 148-50, 166, 174, 19B-202, 203-207, 224-25,229-30. 232, 248-49, 2rf6, 334-35. 343-44, 347, 348, 349, 351, 362, 364-65 Hialoglisdr Scaccario, 81 Diaphragm, 14 riiefceni, Charles, 360 Dictionaries: debates over. 95-99 definitions in, 33, 164-67 history of, 164-67 prestige of, 95-99 recovery of meaning from, 164-67 in transformational grammars. 156-59, 21В.22П, 225-26 usage labels, 98-99, 166 Dictitiintry uf A mericair Regional Easilifl. 1(16

414

Subject Index

Dictionary of the English language, 165, 205 Dictionary of the English Lanx/tage Compiled for the Use ot C Schools in the United Stales, A. 364 Diphthongs: Middle F i n i s h . 340 Modern English, 144, 309, 310-11 Old English, 327 Displacement, 16\ 18 Dissertations on the English Language, 105 Do: anaphoric, 272 causative 272, empty, 273 inlcrroj-jtive, 28J negative, 279 Domesday survey, 69 DutitT Francois, 7fi Double letters, 316,331-32, 362-63, 364 Double negative, 96, 2SD-81 Dowes. Henry, 93 Dravidian, bo now ings from, 102 Dream, of the Rood, 170 Drift, 188-92 causes of, 190 Dryden, John, 34S Ductor in Lingttas, I &4 Dutch, 22, 45. 4S, &* borj'owinps frarci, 101 Eadwig, 59 Early Modern English: development of social dtaiecls, 92-94 extertiallti$tory( 92-110 grammar at 247, 24% 250-51.252, 262-65, 273-74, 280-S1. 2B4-85 phonology tif, 342-52 East Anpiia, 55.5u.5S East. Germanic, 48 East Midlands: dialed. 64, 69, 86 economic growth оГ. 69 emigration from, 69 Eastern New England, 103, 358 Echoic words, 120. 142-44 , 59 Athelin{|, 6d Edmund Ironside. (i2 Edward Г, 71,72, &S Edward 11.71.77 Edward 1ГТ, 72,77 Edward rhe Oanfessor, 65

Edward the Elder, 59 Efik. 360 Egypt, 42, 312 Egyptian, И Eleanor of Provence, 70 Elevalion, 19B Elizabeth I, 349 Eiyot, Sir Thomas, 93 Emma, 65 English; in relaiian to other lajigujtc;^, 4в see also Early Modern English, Middle English, Modem English. Old Enslisb, etc. English Dtctivtutrii.', The, 123 English Grammar, 94 English Expositor. 164 Eiteyttnx, 92 Erasmus, 91 Essays, i 17 Essex, 55, 56 Ethandun, Battle erf, 59 Ethel red the Rcdefcss, 62, 65 Euphemism, 106, 186-87, lS«-203 causes of, 202-203 Ewe, 360 External history, 6 Faeroese, 4B False etymology, 31R Feminine pronouns, 243, 299 Fielding. Henry, 170 Finnish, 45, 46 Finno-Ugric. 46 Fit, of sound and spelling, 301 FitzStephen, William, 81 Flemish. 48 borrowing, 101 Folk etymolagy, 192 Franklin, Benjamin, 364 Fraternitye of Vognbnndcs, 205 Freeman, E. A., SO French: attitudes toward, 64, 71-86 borrowings, 67, 68, 81-83, 90, 92, 135-36, 149-50, i 61 efTect on Modern Enelish, 67-68, 81-83 imposition of, 66, 83-86 Frisian, 6, 48 borrowingK, 101 Front vowels, 144, 309 Function w»rdi, 163 Functional Word Order in Old English Subject-Object Pattern, 254 Fufiare, 315

Subject Г rules Стае] i с, 48 Gallic, 4S G eners t i ve grammar: 2 S~29; Deep structure; Surface structure Gender, 23 £-39, 245 and pronouns, 2+3-44 Genitive: group, 252 inflected, 234-35, 252-53 in Middle English, 252-53 jri Modern English, 252-54 periphrasis, 253-54 in Old English, 234-35 German, 4S. 68 Germanic, 48-51, 110, 167 Ghana, 107 Gill, Alexander, 94, 345, 364 Glide, in diphthong, 30? Gla.isograpkia, 165 ffiotta], 30S Gloltrs, 14 Gloucester, 69 Gloucester, Duke itf, 77 Gothic, 4*, 167 Goihs, 50 Grammar, 23-32, 215-99 change: adjectives, 357-58 adverbs, 190-91. 258 impersonal, 289-У0 negalive, 280-fll noun phrase. 245-54 questions, 284-86 verb phrase, 259-66, 270-74, 280-81, 284- 86, 239-90 causes of, 254-56 descriptive. 22, 26-28 generative, 28-32 traditional, 24-26 and usage, 28-29, 95-98 Great Seal, 73 Great Vowel Shift, 343-48, 354 Greek, 45, 68, 91, HO, 167,312, 314 borrowings frum, 92 Greene, Robert, 205 Grccnough, James, 204 Grimm's Law, 320 Grosxeicstc. Robert, 79 Group genitive, 252 Guinea, 107 Gullah, Ш Gypsy, 47 Haardraada, Harold, 66 Haitian Creole, 3uO Hamito-Semitic, 46 Hamlet, 23, 249

Handbook oi rhe Lixguitric Geography oj New England, J06 Ha::ctfyng Synne, 75 Нагтал, Thoinas. 204 Harold 11, 6b Hart, John, 93,364 Hastings, 66 Hausa, 360 Havetok the Dane, 74 Hebrew, 46 borrowings from, 102 Hellenic, 47,49 Henry 1, 70, K0 Нету Н, 70, 7Я Henry HI, 70, 72, B5 Henry LV, 71, 72, 76, 77, 7S, 80, 65. 86 Henry V, 71 Hermygenes, 42 Herodnlns, 91 Higden, UatuiLph. 79, 80, 85,93 High German, bonowings from, 101 High voweli, 144, 309 Hindi, 47 borrowing*, (02 Illstorhi Eccltsiasika Gentfs Anglorum, 52-53 History, Study of, 3-10 History of the Norman

Co>ti}Ui.'itt S3

Hittite, 48 Hodges, Richard, 364 Holniej. Oliver Wendell. 204 Homer, 23,91,204 Hopkins, Gerard Manley, 183 Horace, 204 Hundred Years War, 71 Hungarian, 45, 4(5 Tfye Waye to she Spyttel House, 205

Icelandic, 4b Ideographs, 3 U Immigration: into London, 63, 69, 86 to the Unired States, 104, 357 Imperative, 266 Impersonal, loss of, 289-90 Indian !a.njju;ij:es, by rowings from- 102 p cognate words, 44. 45 r 46.47, 320 culture of, 4fi homeland of. 46—47 In Лес lions, пудт phrase: Los* of, 254-56 Middle English, 245-49, 252-56 Modern English, 124-25.252-53 case, 234-35, 23JS

41л

416

Subject Indc*

Inflections, noun phrase {coiu.) Modern Fni-lish (com.) number, 239-41 Old English: Case, 2J 5-38 Gender, 238-39 Number, 2 4 1 ^ 2 survivals from Olrf English, 256-57 Inflections, vtrb phrase: Perfect, 260-62 Person, 262-65 Preterit, 260-62 Progressive, 265-66 Transformation rules for, 269-74 Information m Those ItVpo W&dd Remove to America, 108 insular script, 315 Intensifies, L9O-91 Interdental, 30* Internal history, 6 reconstruction in phonology, 321-24 Intonation contour, 27, 295-97 Intransitive verbs, 215-22 Irish, 55 Irish Gaelic, 48 Irregular verbs, 260-62, 290 Italian, 48, 68, 167 borrowings, from, 100 Italic 47,4?, 314 Ivory Coast, 107 Japanese, 45 borrowings from, 102 Jargon, 203 Jespersen, Olln, 115 loan, 76 lohn, King, 70, S3 b h n of Salisbury. 73, 143 John pf TYevisa. 93 Johnson, Samuel, 95,97, 150, 165, 206, 364 dictionary, 95,97, 165 spelling, 364 usage. 95,97, 165 Jones, Daniel. 367 Julius. Caesar, 51,91 Julius Caesar, 16? Jutts, Si, 54 lutish, 6 Juvenal, 204 Kenriclt, William, 165 Kent. 53,54, 55,56, 63 Christ^nization of, 56 place names. 54 Kentish, 55

Kinesics, 163 Kingston, Rkbart), 76 Kinship vocabulary: Modern English, 34, 35 Old English, 60 Killretlge, George, 204 Knighthood, 80, 85 Labio-denlal, ЗОВ Language: change, 36-39 evolution o(, 11-20 design features of, 15-16 description of, 20-32 and meaning, 32-36 and ihyught, 32-35, 160-61 see alsu Bee; Primate Laryn*, 14 Lateral, ЗОВ Latin. 11, 12, 20,22, 24, 25,44,45, 48, .10, 67, 6B, 71, 72, 73, «7, 97, U0, 115, 167,312 borrowings from, 50, 57, 67, 87, П prestige of, 11, 12, 2(1, 24 h 25, &7-S9, ?1^2,95, 138 writing, 314 Laynmon, 74 Lrxendof Good Women. The, 297 Letters Patent, 72, 73 Lettish, 47 Lexical field: *ee Semantic field Lexicon, durability of, 41, J3 Liberia, 107 Life of Fisher, Tiff, 342 I.ily. William. 343 Limb ic system, 14 Lincoln, <59 Lingua franca, 107 Lithuanian, 47 T.ivy,91 Loan translations, 116 Loan words: nee Borrowing fca, 94

Lombards, 50 London, 69, 70, S J dialect of, S6, &2, 105, 344-45, 357 history of, 63-64 Long consonants, 331 Long vowels, [44, 309 Early Modern English, 343-48 Middle English, 336-40 Modern English, 311 Old English, 325-27 Low (Tierman, 45 horrowines from, 101 Low vowels, H4 7 309

Subject Index Lowth, Bishop Robert, $$t 91, 281 Lungs, ] 4 Luther, Martin, 91 Macbeth, 135. 169, 353 Magna C a m , 70 Malay-Polynesian, borrowings from, 102 Manx, 4S Marker, in spelling, 316 Meaning, 32-36, 15Э-Ш as categories, 33, 155 change of, 170-211 defined,32-33 and dictionary, 53-35, 164-67 as distribution, 156 encyclopedic, Ш-Я5 global, IS I as Image, 156 linguistic 1&2-85 in linguistics, 156 as mediation, 202 in neurology, 32 in phiiosophy, 33, 156 in psychology, 32, 154-55, 202-203 radiation of, 171 rcconstruction of, 164-70 and semantic differential, 155-56 semantic features, 34, 156-59, 171, 174 as stimulus-response, 155 structure of,33, 34, J 5t>-59, 161-64 theories of, 32-33, 154-59 as truth condition, 156 as warJ association, 156 Men, pejorative names for, 197 Mercers Guild, 72 Mereliant of Venice, The, 24? Merda, 53, 55,56,59, 63, 64 Mercian, 56, 64 Metaphor, 170, 176-35 causes for, 179-82 as euphemism, 201 and slang, 2Q4-207 structure of, ] 7^—83 Method?, 93 Metonymy, 1EB Mid voxels, 144, 309 Middle Enqlish; dialects, 245-17, 263-65 external history of, 65-90 grammar of, 245^8, 251-58, 263-66. 271, 27^-73, 2SO-SI, 2S4-35 plionology Of, 335-4^ Midlands: from, 46. 247

417

Midlands (cant.) wealth of, 55 Miihiuiimer Night's Dream, A, 30 Milton, John, 5 Minsheu, John, 164, 165 Mistakes, as meaning cbanit, 193 spelling, 321-23 Modal verbs, 2Ы-П Modern English: dialects, 103-109,357-61 external history of, 91-110 grammar of, 120-39, 151, 215-27 229-30, 233^5, 249-50, 252, 256-57, 259-70, 274-77, 27 8~S0, 281-84 phonology of, 306-12, 357-бв Morville, Helewisia deh 78 Morvffle, Hu^e Jc, 7S MuJcasler, Richard, 364 Murray, Gilbert. 367 Names: Norman, SO words from, 136-38 NanocephaLic dwarf, 13 Narrowing, 170-75 causes of, 17.1-7 \ >Tasal, ЪШ Native inheritance, 119 Negation: change of, 2Я5 double, 96, 280-81 Middle FnpMsh, 2Я0-8! Modem English. 27E-30 Old English, 2S0-81 New and Complete Dictionary of the English Language, Thr, J66 New Dictionary, 165 .V4>H.' Wotidol Words. 165 New York Review oi Books, !79 Non-vocalic sounds, 307 Norman Conquest, 63-66 demographic соп$ео.испсе.ч of. 6869, 84 Norman French, B3-86 adoption of names, 80 attitudes toward, 72-SI influence on spelling, 315-16 use of, 72-SI versus Parisian French, 74, 76, 78, 79, 80 Norsemen, 50 North Ocrrnanic, 4S Sforthern British Dialects: at fit tides Inward, 92-94 influence of, 24S-4S, 265-64, 273, 274 Northhampton, 69

418

Subject Index

Northumbria, 53, 55, 56, 58 Northumbrian, 55 Norwegian, 48 Norwich, 69 Noon, 22^-57 compound words, 120-24 conversion, 134-36 count versus non-count, 226-27, definition of, 24-25, 27 derivation of. 124-132 inflection*, 229^*9, 252-54, 254-5B irregular, 257 Noun phrase; transformational rules for, 236-37, 240-41, 252 ste also Syntactic changes Nucleus, of diphthong, 309 Number: Middle English, 245^9 Modem English, 239-45, 256 О id BngHsfa, 239-45 pronouns, 234-35, 238-39, 243-44 Odin, 50 Old English: dialects, 52-56 external history, 52-64 grammar. 121-23, 125-31, 230-33, 235^5. 25 4-5 6, 260-67, ШB1, 2B4, 289-90 origin of, 52-54 phonology of, 324-32 vocabulary, 43, 50, 54, 57, SB, 61, 160 writing, 315-16, 324-25, 327 Oluf, King. 66 Old French, 167 Old Irish, 167 OLd N o w , 167 Old Prussian, 47 Old Slavic, 47 Openness, 16, 18 Origins, 110 Orm, 362 Ornmliini, 362-63 Orton, Harold, 367 Омап, 4& Overcompeiuation, S4, 149, 346-48, 349 Ovid, 9) CM and the Nightingale, The, 74 Oxford, 69 Oxford, Carl of, 95 Oxford English Dictionary. 99, 109. 134, 143h 166,211,215 Paralinguistics, 23, [63 Paris, 78, BO, 85. &6

Paris. Matthew, SO Part of speech, defined, ^4-25, 27, 2S Particle, 28G-&S) Passive, 270-74 with progressive, 273-74 Past tense, 260 Psati, The. 75 Peasants' Revolt. 71, 85 Pccham. Bishop, 79 Pejoration, 19B Peppcrers Guild, 72 Perfect, 260,271-74 Peripheral speech mechanism, 14-15 Persian, 22, 47 borrowings from, 102 Person endings, 262-65 Middle English, 263-64 Old English, 263-64 Peter borough Anglo-Saxon Chronicle, 71 Pelronius, 204 Peveasty. 6tf Phiiadeiphia, 1(13 Pliilip, Kins of France, 70, ЯЗ Phillips, Edward, 165 Pboeniciaii, 314 Phoneme, 303-306 Phmesfheme, 147 Phonetic distoftion, as c^Lphemism, 203 Phonology: change: in Early Modem English, 342-52 in Indo-European, 320-21 in Middle English, 335-^(2 in Oid English, 325-32 causes of, 354-57 reconstruction of, 31R-24 and writing, 316-17 physiology of, 307 set also Sound change Phrygian, 11 Pictish, 48 Fictographs, 312 Picts, 51,52 Pidmn languages, 107, 108, 360 PfeM Pfowwien, 75 r 76 Pitch, 295-97 Plato, ? Piaudeutsch, borrowing, 101 Plautuj, 204 Plutarch. 91 Poiters, 71 Polish, 45, 47 Po, 1ychwл icon, 19 Pope, Alexander, 323 Population, post-1066, 6Я-69, B4 Portuguese, 50, 107, 360 Preface to Shakespeare, 150

Subject Index 41Э , history of, 125-2В Prepositions, 237, 286-89 Prescnplmsm, 5*5-96 Preterit present verbs, 264 Price, Owen, 94 Priestly, Joseph, 97 Primates: hmin structure, 13-15 language of, 15,17, 38 Printing, 6, 36-36,91,317 Privy Seal, 72, 73 Progressive, 265, 273-74, 299 with passive, 274 Pronouns: gender, 243-44 Middle English, 245^9 Modern Fnutish, 234, 23Я-39, 244 Old English, 232, 235, 23S-39, 24243, 299 yvti-lhoti contrast, 249-49 Pronunciation of English in the Atlantic Stales, 106 Proposed for Correcting, Improving ttmt Ascertaining The EnglLili Tongue, A, 95 Provence, 167 ProviiLoriK [>f Oxford, 72 Psammetichos, Kin;;, 11 Psycho-cultural words, 137, 195 Psychology euphemism, 202-203 meaning, 32, 154-55, 202-203 Pultenham, George, 93 Home Dictionary of the English LvngtioRg, The, 69, 99 Rebus, 3)3 Reconstruction, 22, 25, 27, 29, 30, 35, 46, 49-5 L Of culture, 43-47, 48, 49-51, 54-55, 57-5», 61, 81-82, 90 of meaning, 167-70 of phonobsy, 318-24 Relations among languages, 43-4? Restive clauKes, 249-51 Rctroflex, ЗОЯ Reverse spelling, 322, 323, 335, 344, 350 Rhymes, evidence from, 322, 323, 345, 3 if-,

Richarel Г, 70, 71, 74

Richard П, 11, 76, 77, 80, 85 Right Spelling, 94 R Libert of Gloucester, 79

Roget's Tfxr.wurm, 146, 210 Romance languages, 22, 47, 1*7 Romance of Richard the Lian-HearUd, The, 74

Romany, 47 borrowings, 102 Roosevelt, Theodore, 367 Root creations, 120, 142-47 Koot modification, 13?—42 Ruiwalsh, 204 Royal Society, 94 Rudiments of Ertglixh Grammar, The, 97 Rule-governed behavior, 119 Rumanian, 45, 4B Runes, 315 Russian, 45, 47 Ruth we II Cross, 315 Questions: Middle English, 2B4-S6 Modern English, 281-86 Old English, 284-86 5a iL:, Robert L., 254 5a[lust, <>l Samson, Abhot, 7B Sanskrit, 12, 20, 22, 44, 45 : 46, 47, 320 borrowings, 102 Santayana, George, Л Salem languages, 47—i8 Saxons, 52, 53, 54. 63 Schtrut for a New Alphabet und Reformed Made of Spatting, Л, 363 l r , 94 Scots Gaelic, 50 Scottish, 55, 167 Svaiare, The, 56 Semantics: change, types of, 170-93 families, 110 features, 34, l5<5-5<>, 171Л74 fields:

animals and food, 90 color words, 160-61 female reference, 136-97 kinship, 34, 5% language terms, 38 male reference, 197-9B obscenity, 20L occupations, 90 sensory words, 20^-11 laws: abstract words, 20& color words, 208 sensory words, 209-11 replacement. 193 structure. 33, 34. 156-55», 161-64 Semiticwritinu, J13—14 Semi-vowel, 30S Seneca. 91 Scnlac, 66 Sensory wo]4ls, development of, 209

420

Subject Index

Sound chunges in ttiglish icon/-) Sentence: Middle English: defined, 25, 27 /Jong vowel/ > /short vowel/, 336 parts of, defined, 24, 2.4, 27, 28 /short vowel/ > /long vowel/, 337 Serbo-Croatian, 47 / i / > / * / , 337 Shakespeare, William, 5, 30, 169, 24?, / e / > / a / , 337 323, 349, 350, 35 L / u / > S/, 337 Shaw, George Bernard, 360, 367 / o / > / 5 / , 337 Sherwood, Robert, 325 /№/'> /»/, 337 Shift, 1Я5-8Я /a/ > / /, 337 cause* of, 1 -BS /a/ > /ft/, 34В-4У euphemism, 200 M > A 340 Short introduction to English Grammar, [J] > 0. 340 A, 95, 281 LTI > Л 340 Short vowels: /i/ > /Ы/, 343^t4 Early Modern English. 34&-50 /п/ > /эй/, Л43-44 Middle Bngiiab, 336 J O /t/ > /I/, 344 Modern English. 309-10 /О/ > /п/, 344 Old English, 324-27 /S/ > /Ё/, Э45-4К Shorten ins, phrases, 140 /a/ > / * / , 345^18 Shortening, vowels, 336—3S An-/, /gn-/, /hn-/, /hl-/,/hr-/ > Sierra Leone, 107 /к-/, /g-ЛЬ-Л 333 Simplified Spelling Board, 367 Early Modfirn English: Sir Gawtdn and rite Grern Knight, 75, 249 /r>i/ > /ai/, 343-J4 Ske;U, Walter, 367 /im/ > /аи/, 343-44 Slang, 204-207 /Ё/ > /I/, 345^tS Slang: Yesterday ami Today, 205 /3/ > /6/, 344-45 Slave trade, 107 /a>/ > /e/, 345--(B Slavic languages, 22, 47, 4? /a/ > /S/, 348-49 borrowings from, 101 /г/ > /a/. 348-19 Slovenian, 47 /N/ > /n/ - /rj/, 350-51 Social dialects: see Dialects, social /-И + У - / > /z/,351 /vowel + r/ > /vowel + 0/, 352 Sodele de Linguistique de Parts, 12 Sound symbolism, 142-47 Socrates, 42 Spenieh. 22, 45, 47, 50, 68, 107 Sound change: borrowings front, 100 assimilation, 355 Speculum Vitae, 15 causes of, 354—57 Spelling 105,312-16 diuiaiilatien, 355 British-American differences, 363-65 functional load, 354 consistency, 317 GrsaL Vowel Shift, 343-4В, 354 vowel lengtheninB-shortenine, 336-3E conventions of established, 315-17 tec also Early Modern English, Middle Middle Pni-lish vowels, 335 English, Nfodern Enfiiish, Old mornhogrnphemic, 366-68 English, Phonology morphophonemlc, 366-68 Sound changes in Fngiish: Old Englkb., 315-16, 324, 326-29, Old English: 330-32 /B/ > /ft/, 326 and pronunciation, 301 /h > /i/, 32* reform, 362-68 A / > 1Ы - /k/, 328 objections to, 365 /sk-/ > /s/, 32Я reverse, 322, 323, 335, 344, 350 /g/ > /J/, 3J0 and Noah Webster, 363-64 /a/ > /Г/, 335-36 Spenser, Edmund, J70 /spirants/ > /voiced/ — Spirants, 307 si, Augustine, 56, 9) /voiceless/, 330-32 Statutes of Parliament, 73 /Loft£ conson&nW > /short Stephen, Kiflfc 70, SO, R5 tonionants/, ?31-32

'

Subject Index Stern, Gust;Lf, 20& Stops, ЗО7 Stress, 295-99 Style: borrowed words, 61, S7-S9, 99, 106, 112-!!, 131-33, 147-50 compounding, В9. 121, 123, 148 levels of, 21, 62, 143, 149 linguisljc change. 84, 148-49, 324-25 set aba Direct, style Subjunctive, 266 Suffixes, 124-33 derivational, 124-33 inflecLionai, 124-25 Midttlc English, 245-<7, 252-53, 25Я, 262-66 Modern English, 124-25, 227, 233-

35, 239-41.258, Z5?-«

Old English, 23 3-45, 258, 25!>-66 Л?г iif.Tfl InfleclmnK Sumerian wriling, 313 Sirprasegmenlals, 23, 245-99 Surface structure, 31, 215-21 Survey of Verb Farms in the Eastern United States, A, 106 Sussex, 55, 56 Swahili, 45 Swedish, 4^, 4 * Swifi, Jonalhiin, 95 Syllabic writing, 313 Symbol, in spelling. 316 Synesthrtic swirid symbolism, 144 Synesthesia, 209-11 Syncope, 140 Syntadic change, summary of, 290-91 Syntactic chunpts: adjectives, 244^15, 258 adverbs, 2-5S articles, 22*, 295 atuiliury verbs, 270-74 comparative, 25S direct [jbjrcts, 221-25 jenitive, 252-54 impersonal verbs, 289-90 Joss of dative, 246, 255-56 loss of thee, 249 negative, 280-S1 rrepositional phrasei, 23fl, 252-54, 287-89 ргопоиля, 99, 246-49 question, 284-S5 relative d a u « . 2J9-52 subject, 221-25 SUperlutive, I^« verb + particle, 2Й7-89

421

Syntactic changes (cvttl,) word order, major elemenls, 230-33, 254-56 you, polite, 24S-49 Tacitus, 91 Tag uucsiions, 277 Tamil, borrowing, 102 Teeth, J4 Telej^u, borrowing, 102 Terence, 91 Thackerv, Williarr, 360 Theocritus, 91 Thin^, 62. Third person, source of, 263 Thur, 50 Thuuydides, 91 Tibeto-Chinese, borrowings /rum, 102 Tocharian, 47 Ta»t Jonrs. 170 Топкие, 14 Traditional grammar: origin of, 24, 94 and prestripiivis [henry of, 24-25 Transfei1, 170, euphemism. 201 obst-enily, 201 see aha Metaphor Transforrnalions.: хее Crantmaiical change, Syniatiic changes Transitive verbs, 215-20 Trees, grammatical, 218 Trees, family of Indo-European languages, 49 Trevisa, John, 80 Tryon, Thomas, J70 Turkic, borrowings from, J02 'I itrkish, 45 Twelfth Night, 24* Twi, 360 Two Gentlemen of Verona, The, 249 Ulfilas, Bishop, 48 Ulster, migration from, 103, 35& Umbrian, 48 U m W l , 257, 294 I MI^': attitudes towarJ American English, 104, 364-65 insecurity, ]()(j lar verbs, 261 derivations, 132 reconstruction of, 29 See aho nialects, Sficial, Ktyle sk, Thomas, 76

422

Subject Index

Vaadtia. JO Velar, 3Ofi Venus and Adonis, 330

Vert:

causative, 222-26 classes, 260-62 conversion. 134-36 definition of, 24, 26-27 impersonal, 289-90 inflections, 259-77 irttransilive, 2L5-26 irregular, 26(1-61 partkle, 2K6-S9 preieril present, 264 transitive. 215-26 Verb phrase: transformation, 220, 223, 231-32, 266, 269-70, 275, 278-E6 see alsi> Syntactic changes Vcrner4 Law, 321 Vikings, SO, 51 Vilalis, Odericus, 78 Vuciibtitary; s?e Borrowing. Native inherit u nee. Word formal ion Vi'cai Organ, f>4 Vocalic siiundv 307 Voiced sounds, 30* Voiceleu stmnds, зол VoweU: Early Modern English, 3J3-J0 MiJdle Elfish, .1.1.W0 Modern tngli^h. 509-12 Old En^lish. 324-27 symboJism, p g Middle Eji Vulgar Latin, 204

Walter of Bibbesworlh, 79 Watsdertr, 56 IfrtV/fj Health. Thf. 170 W;LI Tyler rebellion, 77 Webber, NuiLh, 105, 166,364 Wtbstei's New World Dictionary ol th American Language, 99 Webster's Second New Intel national Dictionary, 97, 9«, 9V Wc/j.nff;ih,t 77«>f/ Л/гм1 huermiiiotial Dictionary, 67, 97. 98, 99, 166 Wed more. Treaty of, «I Welsh, 4S borrowings from, ?5 Wessex. 55, S6, 58, W, 6!, 64 West African l«n£LiaRts, 360 cognate*, 4S Wesl Germanic, 6, cognates. 48

Wesl Saxon, 56, 5S, 63, fi4 WMUnimter, 63 Whot it History, 5 Whitney, W, D., 3*7 Widening 170, 175-77 СШй of, 177 eupJicmism, 200 Wilkins, John, 364 William the Conqueror, 63, 66, 69, 70, 71,78,80 William 11,70 William of Canterbury, 75 William of Mtlmnbury, 92 William of Naisyflftoo, 75 Willium of Wetimintier, 19 Wlocheiter, 63, 6? Witan, 66 Women, peforativ* names for. \96 Worcester, 69 Word families, 110 Word formation: lK, 141-42 formiitioa, 13.1-54 Mends. 140^1 clipping, [39-40 compounding, 120-24 tonvcrsion, 134—36 derivation, 124-33 from names, J36-3B root crealion, 142-47 root codification, 13P—42 shortening. 139-40 Word Gtogrnpby of the Eastern United Steles. A, 10* Word order: change* in, 230-3? major dements, 229-33 Words; disappenrancc of. 43 origin of, 42 ИС al.it) U[]rrnfwirtj!, Meaning, .Semantic fields. Word formntiofl Wordtwortb, William. 323 Wreck iij the Dcitl-wMiuvh The, I S3 Writing, history of, 312-16 evidente Tfwm, 31Я-24 prgstiyc []f, 21 ife a ls